There was a spurt of debate on Twitter today about what is GIS. Specifically, if many of the new innovations around location are all just re-skinned GIS or if it is something fundamentally different. The divide seems to be roughly between folks that are long time GIS users and folks new to space or have left GIS disillusioned. The long time GIS users see new applications as simply being slight variations on what they’ve been doing for decades. While the new entrants see their work as something new and different from GIS and its legacy.
There is a case to be made that a Geographic Information System covers any technology that is expressing location digitally. Wikipedia gives almost as broad a definition “GIS is a system that captures, stores, analyzes, manages and presents data with reference to geographic location data”. So, the GIS clan on Twitter has a point, no doubt. Then why do I disagree so vociferously against it? The big reason is from reading/witnessing decades of literature and positioning by the GIS community that is more than just a tool or points on a map. That GIS is a science for trained specialists to employ. Now that the tide is turning in the opposite direction we want to conveniently redefine what “GIS” is to suit the next market opportunity. For those of us who purposefully left the path of GIS and science/specialization it seems a bit odd.
For those not familiar with the history there was a debate in the early nineties about the future of GIS and what it should be. On one side you had the seminal paper by Mike Goodchild (1992) stating GIS was more than a tool but a science.
…the development and use of theories, methods, technology, and data for understanding geographic processes, relationships, and patterns. The transformation of geographic data into useful information is central to geographic information science (UCGIS 2002).
Practitioners on the “tool” side of the debate thought GIS should be concerned with advancing the technology’s capabilities and ease of use (Wright et al. 1995). In contrast the “science” side believed that without a focus on the academic specializations in geography as a discipline, “(GIS) is likely to degenerate to data management and map making, however complex the tool’s capabilities for scientific analysis and modeling” (Wright et al. p.335 1995).
The degeneration feared in the nineties is exactly what has happened today. The simplification of tools to bring geographic data to an entirely new audience. This is exactly what the definitions and taxonomies of the nineties wanted to avoid because it meant GIS would end up in computer science department and not geography department. GIS would become part of the IT main stream. Even now the same GIS establishment fights to differentiate the old from the new, labeling data not being generated by GIS professional as “volunteered geographic information”. In short geographic data generated outside of a GIS must be volunteered to it and verified by GIS professionals.
Yet we are debating that these same technologies are actually a GIS as well. You can’t have your cake and eat it to. Are they equal GIS systems or subservient to the GIS that professionals use. It is all a big contradiction. The reality is the GIS establishment will create whatever definition is convenient to protect the livelihood of GIS professionals and the business of companies who supply them technology. That is their prerogative, but I do not think it is accurate to lump the innovation happening outside that establishment as being part of their collective efforts. Whether you call that innovation NoGIS, NeoGeography, or NextGIS is superfluous to the fact it is a distinct fork in the road and fundamentally different from the GIS establishment that has created “Geographic Information Science” and “Volunteered Geographic Information”.
12 Responses to The Contradictions in Defining GIS
Leave a Reply
New GeoCommons Maps- KIN ZIP dboozer
- UKPoliceTaserUse indeuppal
- Connecticut Pedestrian Fatalities with Rail Buffer RenataPS
- TrainCrimesUK2011 indeuppal
- February 21-22, 2012 CO & WY High Wind Event ExaminerWeather
- Untitled Map reversededgesword
The Science behind GIS is called Geography… GIS is itself not a science.
from Dictionary.com:
ge·og·ra·phy
[jee-og-ruh-fee]
–noun, plural -phies.
1.
the science dealing with the areal differentiation of the earth’s surface, as shown in the character, arrangement, and interrelations over the world of such elements as climate, elevation, soil, vegetation, population, land use, industries, or states, and of the unit areas formed by the complex of these individual elements.
Just the cycle of innovation — from the first airplanes to today’s gps guided auto-pilot; from the first computers when you had to know how it worked to the folks now that have no idea why their smartphone is slow as they are running too many apps; from the secretary to word processor to Google Docs.
Maybe it is people’s altered or inflated definition of “information” and “system”.
There will always be the scientist that has a pure view of their work; and non-scientists that find common nature/sense out of that pureness.
People want to know they are doing “God’s Work” (so to speak) and that it is somehow important; especially those that define themselves in their work.
It is hard to let that importance be undervalued as it may be perceived as undervaluing the individual.
GIS is simply like Money? Different forms over the years (gold, coin, check, plastic, wired, but still the same in the end); the value is harder to justify year after year with the same effort (inflation); that analysis pieces are always trying to predict and trend it but now has to deal with the vastness of more data in different forms (economics).
As the wikipedia definition suggests, there are so many aspects to this discussion. Its funny that it is tagged as a title=”peacock term”>
At the risk of sounding like an obscure geographer, its interesting to note how “GIS” is mixed up with culture. For example, the French and Spanish definitions of GIS on wikipedia don’t refer to GIS as a “science,” unless, in the spanish case, they actually quote the english academic name in english. The French definition clearly puts it in the category of simple information management. Maybe this affirms, as you say, there was an anglophone academic movement for GIS to be defined as a science. What have we gained from this movement? Why did it happen? Interesting questions, thanks for sharing your thoughts.
ha, oops
You had some nice points here. I done a research on the topic and got most peoples will agree with you
Good summary Sean, I think you captured and clarified it well.
Twitter memes and tongue-in-cheek labels aside, there is an increasingly large amount of data that no longer fits into the feature (point, line, polygon, layer) data model that current practitioners of GIS use. Data now has both temporal and spatial dimension and frequently relationships or linkages to other data. For example, sensor feeds, twitter feeds, activity streams, and Reality Markup Language have these attributes. While this type of data is well known, see Peuquet</a/ and Yuan, the tools to build a workable data model and implementing it in software have not been available until relatively recently. The availability of schema-less NoSQL datas and on-demand compute clusters makes it possible to perform the analytical use cases outlined by Yuan.
[...] The Contradictions in Defining GIS [...]
I think that the various notions and implementations of GIS have been and are technology (tool) dependent. The 1950′s saw GIS developed on 2 million dollar IBM machines that were, by definition, expert driven and esoteric. The 1980′s and 90′s were defined by the desktop computing era that saw GIS implementation ‘at the grass roots’ with the ‘x’GIS like PPGIS, PGIS, CGIS, etc. 2005, in my mind, was a watershed moment with the launch of GE, which has now been activated over 800 million unique times, and loaded onto mobile devises more powerful than the original IBM. As such, the debate should not be about traditional to tools of GIS and Neogeography; authorized data feeds and VGI. The debate should be about power and empowerment.
PPGIS was rooted in notions of ‘grassroots’ as is Neogeography. The gap that is not explicitly addressed in this discussion is the inherent cost (thousands for a desktop and ArcGIS licence, vs hundreds for an iPhone and data plan or for a netbook) that acts as a barrier for many people. In addition to cost is knowledge. Neogeography IS expert driven. Whoever coded Ushahidi had expert knowledge. Sure I can add data to it, but I cannot create a platform.
For me, a reduction of barriers is important. How can we fully integrate flip phones (most common phone on earth) into the neogeo phenom? Is it possible to build open source modular mapping systems that such that someone can deploy a platform that addresses their specific needs with no expert knowledge of computing? History tells us that the cost barrier will be further reduced – we need to meet that challenge with the tools to support grassroots involvement.
It is an old axiom in geography that individuals are experts of their local environments. If that is true we need to deploy the tools to allow the expression of that knowledge, and build an understanding that respects that expression.
Sean, et. al. Thanks for this. Awesome conversation.
I share the feeling of contradiction somewhat, but internally. I began as a surveyor, a lifetime ago, so I get millimeter accuracy and geodesy and that fun world, and projection hell, then I became a GISPerson, and now my team is entering the tiled, script driven world of NeoGIS.
So there’s this establishment, and there’s the creative coder world, but there’s also a middle group of us who live/play in both worlds here. So from that middle group, it seems the desire to stupidly call the field of interest GIScience as well as call the desktop/mainframe applications GISystems is dumb and likely always was a poor call.
It’s fascinating to see what folks at Stamen, SimpleGEO and others all over are doing with space and the web. It’s not yo mammas GIS fo sho, but much of it builds on a core set of ideas and procedures that are solidly in the “establishment camp”. Our shapefiles we build into tiles came from some sorry-ass surveyor and through some enterprise GIS at some point, so we could do much, much more with them than those folks ever imagined. But it would be foolish to try to label those new end tools/outputs as GIS. So let’s not then.
The idea that GIScience folks want to keep GIS in the geography field is not that far from the exclusivity we’re now creating in NoGIS- if you ain’t down with Javascript you cant play in this game. So it’s now coders who own the game? Do we become them?
Then there’s the amazing set of easy to make your own map tools on the interwebs, google, GeoCommons etc, what are they? We could argue they have the elements of an old school GIS fairly convincingly right? But they are gateway tools that opened into this awesome new NoGIS world.
Perhaps a more constructive, positive way forward is for us to establish that GIS will not disappear and is not an all-encompassing field covering everything that results in a map. We owe it a lot, and it owes us some apologies too. Some corporations will try to claim what is being done now and too bad for them because that aint real. We gain more by defining what we are about, not what they are not about though.
Some thoughts from a conflicted, confused GI Something.
Now where’s my theodolite gone…
Mobile is about to have a big impact on the GIS industry. Real estate often proclaim its about “location location location”. Take heed GIS industry. We are moving beyond GIS to location based application development. If we sit in our niche we will miss the boat:
http://www.webmapsolutions.com/mobile-arcgis-application-development