Introduction to Neogeography - From O'Reilly Media
Several years ago the geospatial community was abuzz with talk concerning neogeography and the proliferation of online maps created by amateur cartographers (i.e., web mapping). These conversations often revealed an ambivalence among traditional GIS practitioners who on the one hand welcomed the advent of new tools and data, and the promotion of maps for visual communication, and on the other sought to distinguish their practices and training from the novice.
These tensions were palpable within many small liberal arts institutions where expectations for GIS support and instruction, propelled by the rising popularity of map-making and spatial studies, exceeded those resources available to support these desires. In this new environment, trained GIS educators/users intermingled with ambitious, but less experienced, colleagues who sought a quick foothold in the field. How these new participants and accompanying transformations in mapping technology would impact the perception and practice of GIS on campus remained unclear. Would neogeography dilute traditional cartography or evaporate as yet another fad?
The ultimate response will likely lie somewhere between these outcomes. It strikes me, for example, that online references to neogeography are on the decline. If I am correct, this trend may reflect our growing acceptance of the new mapping democracy and it’s place in popular culture. In academia, campus staff who once sought to support GIS software (e.g. ArcGIS or Manifold) without a long-term commitment to professional training and use increasingly recognize the difficulty of this effort.
Confusion and Questions
It’s revealing then to hear more-and-more undergraduate students refer to web-mapping applications as GIS. Apparently, a number of people do not distinguish between these parallel practices. Instead, those factors unique to traditional GIS software (e.g., integrated data storage, editing, and analysis – particularly analysis) are confused with the more generic use of mapping services applications and API’s to create spatial visualizations on the Internet.
To what extent then have GIS practitioners in higher-education incorporated web-mapping applications in their instructional tool-set? What impact have these practices, and others, had on our understanding of GIS as a term with specific meaning?
While I cannot address these questions with particular authority or certainty, it might be interesting to consider how the popular use of such terms as neogeography, GIS, Google Earth, and Google Maps (the latter two representing popular web-mapping platforms) vary through time and what implications these patterns may have for cartographic professionals.
A Google Trends Analysis
To evaluate such patterns, I queried Google Trends (GT). In use, GT “analyzes a portion of Google web searches to compute how many searches have been done for the terms submitted for evaluation, relative to the total number of searches done on Google over time”. By default, the data are scaled to the average search traffic for a term -represented as 1.0- during the selected time period and are displayed on a Search Volume Index (SVI) graph.
Below are GT visualizations from my search for the term “GIS” in combination with “Google Earth” and “Google Maps” (Figure 1) and “GIS” alone (Figure 2) in the United States from 2004-present. I also searched for “neogeography” and “geographical information systems” but there was insufficient search volume to create a graph for either.
Figure 1. Trends in Search Frequency for "GIS", "Google Earth", and "Google Maps" Within the United States
Figure 2. Trends in Search Frequency for "GIS" Within the United States
In Figure 1, the search terms “Google Earth” (red line) and “Google Maps” (yellow line) have been scaled relative to “GIS” (blue line). We clearly note that subsequent to 2005, when Google Earth and Google Maps were released, the volume of searches for these terms exceeded that for GIS and were frequently 5-10 times that of the former (std. error ~3.0%). The SVI for Google Earth peaked in mid-2005 and then drifted down through early 2006 to about 5.00, where it more or less remains to present. I note that smaller peaks in the Google Earth trend line through time follow major news releases related to the product. The SVI pattern for Google Maps declines from early 2005 to a low in late 2006 and then increases to present, exceeding the SVI for Google Earth since 2008. Focusing on Figure 2, we note that the SVI for GIS has declined throughout the period of 2004-2008. I also note periodic dips in the SVI of this line at the transition between years that I can’t easily explain. Are these departures evidence of real phenomenon or are they artefacts of data collection?
A First-Order Interpretation
It appears that since 2004, Internet users in the United States have searched less frequently for GIS-related sites and/or news during their browsing activity and are spending more time searching for information related to Google Earth and Google Maps. Unfortunately, the data don’t reveal what type of information these users were seeking specifically or what demographic/professional sectors they represented. It would be helpful to know how these SVI data compare if drawn only from the educational sector.
The SVI trends do suggest a growing popular preference for information related to the Google Maps, particularly in the past two years. This may owe, in part, to the fact that Google Maps has an API that allows easy integration into other visualization frameworks and supports synchronous collaboration for map creation in the online environment. While Google Earth is unparalleled as a virtual globe with 3-D rendering, it is designed primarily for desktop use – which may limit its wider appeal. From the SVI data for Google Earth it is nonetheless tempting to infer a steady interest in this application among a core audience of users since 2006.
This analysis does not suggest any correlation between browsing activity and the relative importance of web mapping and GIS in scholarship. Nonetheless, the data do provide some basis to generate questions, predictions and/or hypotheses related to developments in the geospatial industry and spatial practices in the classroom. From the SVI data, we can conjecture whether certain applications or software will assume prominence in the classroom as a consequence of their popularity. For example, will educators defer to Google Maps in their teaching activities if future students indicate a strong preference and familiarity with this application? The possibility of the latter is suggested by the SVI trends. The data might also lead us to understand when/how advances in web mapping technology will inspire innovation for GIS development and visa versa. ESRI’s ArcGIS Online application and the unrelated beta release of GIS Cloud point to this influence.
Implications?
Online mapping applications are easy to use which makes them practical for class projects emphasizing spatial visualization over analysis and in courses focused on subjects other than GIS. In fact, many GIS instructors in higher education now use Google Earth or a web mapping application like Google Maps to teach fundamentals of spatial thinking and to support the bulk of classroom map production. Our Introduction to Geographical Information Systems workshop takes similar advantage of these tools to complement learning activities with ArcGIS.
I do wonder if such technology integration lends to confusion among novices as regards the differences between traditional GIS and web mapping applications. If so, future convergence in the functionality of these technologies will only further blur such distinctions. How then will the general populace distinguish the studied cartographer and those we once referred to as neogeographers?
I am interested to hear how you view and value the impact(s) of neogeography on geospatial teaching and learning – particularly in the liberal arts. Please feel free to share your experience and insights!
Posted on March 15, 2010 at 7:19 am by Sean Connin · Permalink
In: Liberal Education, Pedagogy, Science, Technology · Tagged with: GIS, Google Earth, google maps, neogeography
In: Liberal Education, Pedagogy, Science, Technology · Tagged with: GIS, Google Earth, google maps, neogeography
on March 21, 2010 at 10:13 am
Permalink
Hi, Sean,
Your trends analysis is a very interesting approach to finding out where the mapping field is going. I would suggest including Yahoo Maps, also, as many people have used that technology in preference to Google Maps, for roughly the same length of time. From now on, of course, you’ll need to keep track of Bing Maps and perhaps even ArcGIS Online! This is a rapidly moving field.
My feeling is that on-line tools of many kinds, not just mapping, will continue to expand in their use and capabilities. And online mapping tools will become increasingly analytical (though they still have some work to do with improved display). Anyone teaching in this area will have a continual challenge to be aware of new possibilities, and use the tools that get the job done with the least amount of effort. Some of this challenge will come from their students, who are often more aware of the “next great thing” and will try to put it to use. Sometimes this is great, other times it will be inappropriate for reasons that the student doesn’t recognize. For example (one of my experiences) after teaching students how to use ArcGIS with its panoply of options for cartographic embellishment, a student instead used a Google Map with only its defaults and an orthophoto as a background, producing a very difficult-to-read map (whose intended use was a printed piece of paper!). Another possibility of current concern: using a cloud database and on-line map to store and display sensitive data, an issue for which we need to provide some education.
On the other hand, the ease of generating a map with something like Google Earth or the Google Maps Widget has been a great boon for professors who just want to get a small amount of data on a map. The presence of a mostly good set of default base maps and occasionally useful data, along with GE’s inherent “wow” factor, has been a real boon for the increasing use of geography in the classroom. Many professors have recognized this on their own, and I facilitate their use when I can (even though GE is a relatively simple application, many users are not familiar with all of its capabilities). But I will also encourage their use by professors who are unfamiliar with them, when appropriate. In either case I will point out its limitations and remind them of the better tools we have available, if they need to go in that direction.
I’ll close with my personal quarrel. I would discourage the use of the term neogeography in any application; the term is both time-relative and not really descriptive of the applications it is supposed to encompass: personal mapping, volunteered geographic data, and the social-geospatial web. As the Wikipedia article Sean links to points out, the term has had a number of different uses over the years; it could even have been applied to GIS several decades ago, when it was “new geography”. Perhaps this is why the term doesn’t show up with a sufficient search volume with Google Trends Analysis.
on March 21, 2010 at 10:14 am
Permalink
The dips in the google trends correlate with the end of each year. This is typical of professional and business related searches, because the dips are during the holiday season. This would show that GIS is still a very business oriented term, where Google Maps and Google Earth are used year round for personal use. It even looks like Google Maps queries go up in the summer, probably due to road trips.
on March 21, 2010 at 3:23 pm
Permalink
Hi Sean, When I was looking for more from my GIS/Geography education two years ago, I lucked into a job at a web-mapping company. I ended up nearly failing my GIS classes at university because I was so much more interested in what I was learning at work. So I guess I’m one of those liberal arts students you are referring to.
My own experience was that academia moved way to slowly to address emerging technology. PhD’s are, after all, often a long-term engagement with dinosaurs. Publishing takes forever. I could go on and on. My professors didn’t have the time or incentive to “get it.”
You might enjoy Crowdsourcing by Jeff Howe if you haven’t read it already. He addresses your final question in an elegant way in the first or second chapter.
on March 21, 2010 at 3:26 pm
Permalink
Hi Sean, When I was looking for more from my GIS/Geography education two years ago, I lucked into a job at a web-mapping company. I ended up nearly failing my GIS classes at university because I was so much more interested in what I was learning at work. So I guess I’m one of those liberal arts students you are referring to.
My own experience was that academia moved way too slowly to address emerging technology. PhD’s are, after all, often a long-term engagement with dinosaurs. Publishing takes forever. I could go on and on. My professors didn’t have the time or incentive to “get it.”
You might enjoy Crowdsourcing by Jeff Howe if you haven’t read it already. He addresses your final question in an elegant way in the first or second chapter.
on March 22, 2010 at 5:06 am
Permalink
[...] } Que devient la néogéographie ? Vaste question à laquelle le site National Institute For Technology tente de répondre à travers un billet posté il y a quelques semaines. L’auteur opte pour [...]
on March 22, 2010 at 8:58 am
Permalink
Hi Andy,
Thanks for your thoughtful comments. Interestingly, the SVI for the term Yahoo Maps is very similar to GIS for the period 2004-2010, in terms of both magnitude and direction through time. Far fewer people are seeking information about either Yahoo Maps or GIS as compared to Google Maps and Google Earth.
Both Bing Maps and ArcGIS Online were introduced in 2009 and there was a sufficient SVI to produce graphs in Google Trends. I encourage anyone to generate an SVI trend for these terms and then compare the results. Any predictions as to which application will continue to make a splash?
I also appreciate your thoughts on classroom practice. As you correctly point out, we can expect significant developments in geospatial technology and online environments in next several years. We can look forward to many more conversations like this one.
Sean
on March 22, 2010 at 12:04 pm
Permalink
Tom,
I would argue that it’s the neo people who have yet to “get it”. Google Maps and web mashups are the simplest, crudest kinds of rudimentary GIS. They’re cute but far less useful. You may regret skipping your college courses in actual GIS. When they’ve incorporated all of the different types of analysis and number-crunching that desktop GIS can perform, they can become worthy of respect and inclusion into university programs.
on March 22, 2010 at 3:55 pm
Permalink
Hi Tom,
Thanks for writing in and sharing your perspective. Regardless of how we might champion web-mapping vs. GIS, you brought up an important issue for educators.
The rapid transformation in geospatial applications challenge the best of us to update our skills, let alone our classes. This is an important point considering the often significant investment in time necessary to create a meaningful lesson/lab activity.
In this current environment, there is a very real chance that such work will become outdated within a year or two when built around a particular software/release.
I do think we need to look continually forward to understand how the geospatial profession will evolve in the near and long-term. It’s likely that employers will value skills that weren’t even considered several years ago.
For example, should our undergraduates know how to build a geostack from open source applications? I’ll wager that more than a few of us in higher education would be hard pressed to teach this activity right now.
Thanks for sharing,
Sean
on March 25, 2010 at 12:42 pm
Permalink
Hi Sean,
Interesting, and well written article. As a GIS developer, I have had the good fortune to benefit from both formal GIS instruction in academic settings, and involvement in the so-called “neogeography” tech explosion. I’ve always hated that term by the way, as it smacks of elitism, and generally refers to meaningless distinctions between the 2 fields. My personal opinion is that both are worth knowing.
Formal GIS courses do a good job of teaching people about certain fundamental geographic concepts, like projections and cartographic principles, but fail completely in other regards. That failure is especially apparent with regards to knowledge about data types, and a basic understanding of computational geometry. Simply put, many formal GIS students today seem incapable of solving a simple problems without resorting to some sort of desktop GIS application. Why? My experience is that the analytic process is taught by rote memorization of what buttons need to be pushed in order to solve the problem. In essence, the field seems to be teaching the application, and not the science behind it.
By contrast, “neogeographers” tend to be more results oriented, and seem to be more capable of applying whatever technology is available at solving a problem. Their downfall is that in many cases the only validity check they use is, “does it look right?”. With respect to projections, most only know how to work with geographic coordinates, or in some cases, the spherical mercator projection used in almost all web mapping API’s today. Knowledge of formal cartographic principles, long held by the GIS crowd as being one of the distinguishing marks of a “true professional”, is also generally lacking in the “neogeography” group. Why? Quite simply because Google Maps has shown that it simply doesn’t matter as much as the GIS crowd would like to think it does.
What I wish GIS educators would do, is to use both sides of the coin to teach the profession. Start by teaching basic vector data types and visualization using something like KML , or Geojson in OpenLayers – or even one of the commercial map API’s. This forces people to actually “touch” their data, and to gain an appreciation for the huge advantages provided in many cases by proprietary GIS data formats. Do the same thing with simple cartographic principles. There is something very visceral about editing javascript, reloading a web page, and seeing your changes. You know exactly what you had to change in order to produce that effect. Then move into the desktop GIS application and ramp up on the more compicated analytical and processing functions. And by all means, teach and encourage a basic understanding of geometry, Computer Science, and Web technology in GIS students. There is simply no reason why someone should have to open a desktop GIS application in order to calculate the distance between 2 points – regardless of the map projection. And as much we love plotters and paper maps, our users today want to see the information displayed in their Web browser as well
Finally, both sides of this imaginary fence would do well to eliminate the inflammatory and schism-perpetuating comments that only serve to maintain an imaginary divide. Most of the people “we” work for don’t understand the distinction. To them it’s all just “GIS” or “Maps”, and they don’t care what tool you use to do it.
We should care far less about the tools as well.