We left our “geojargon” survey open for two weeks in March. We asked readers to look at 35 geospatial terms and determine how well they knew the meaning of each. A total of 538 respondents anonymously assigned each term to one of four categories (view results):
- I'm not familiar with this word
- I've seen/heard this word, but don't know what it means
- I think I know what this means, but I'm not sure
- I know the meaning of this word
Well-known Terms
I did some simple analysis, first on the raw data. What were the terms we felt we knew best? The following terms were assigned to the “I know the meaning of this word” category by more than 90% of respondents:
- geocoding
- open source
- spatial analysis
- topology
- schema
- Web service
- raster
- accuracy
The term that surprises me most in that list is open source. My experience has been that many people who think they know what it means actually know only a few of its properties. In fact, Esri just published an article citing misconceptions about open source software in ArcNews. And, as I noted in a recent article about an education conference I attended, technical people developing on top of Google Earth described it as open source, when in fact it is not released under an open source license.
Less Well-known Terms
On the other end of the spectrum, what terms are we least confident we can define? These terms had more than 20% of respondents assign them to the “I'm not familiar with this word” category:
- ontology
- NGAC
- neatline
- NSDI
- conflation
- kriging
- DRG
I am not surprised at ontology being in this group. While used in discussions of standards and metadata, ontology, which refers to a structured organization of knowledge domain, is mostly used in academic discussions, in my experience. NGAC, the National Geospatial Advisory Committee that advises the Department of Interior, and DRG, digital raster graphic, a scanned quad sheet, are very U.S. focused and may not have been familiar to those from other countries.
Digging Deeper
I combined the number of people who selected either of the two “know” categories (“I think I know what this means, but I'm not sure” and “I know the meaning of this word”) for each term. I looked at which terms had more than 95% of respondents select one of those options and came up with more terms respondents felt confident they knew. The terms from this group included the ones noted above (Well-known Terms) plus:
- cloud computing
- choropleth map
- watershed
- parameter
- javascript
- enterprise gis
- data model
I’m skeptical of the high ranking of cloud computing. While I hope most people have a sense of what it means, I’d wager that if I asked respondents to define the term, the responses would vary widely. I suspect that like open source, respondents know of some of its key properties, but not others. And, to be fair, unlike open source, there is no formal definition. (The Open Source Initiative offers this wonderful annotated definition.)
I combined the number of people who selected either of the two “I don’t know” categories (“I'm not familiar with this word” and “I've seen/heard this word, but don't know what it means”) for each term. I looked at which terms had more than 25% of respondents select one of those options and came up with more terms respondents felt less confident they knew. The terms from this group included the ones noted above (Less Well-known Terms) plus:
- blob
- cartogram
- GML
- map algebra
Blob, binary large object, is a set of binary data stored as a single entity in a database field. It is perhaps no longer in widespread use. Now that we have databases that store geodata well, there is far less “blob storage” going on than there was in, say, 1995.
Trends
To try to explain why respondents know some terms and not others, I reviewed the full lists. First, the full list of well-known terms:
- geocoding
- open source
- spatial analysis
- topology
- schema
- Web service
- raster
- accuracy
- cloud computing
- choropleth map
- watershed
- parameter
- javascript
- enterprise gis
- data model
My first thought is that the majority of these terms would be mentioned in an introductory GIS class. The exceptions, open source, javascript and enterprise GIS, are certainly well discussed in the work environment, the geospatial media, at conferences and in online discussions/social media.
Second, some of these terms are core geography and cartography terms; they are not specific to GIS. That suggests that those who came to GIS from a geography background may have run into them before working with or studying the technology. I’m thinking of watershed, accuracy and spatial analysis, for example.
Finally, these terms may also be well represented in any software provider’s marketing materials or documentation, so they might be picked up when looking for software products or taking a training course/self-teaching.
Here’s the list of less well-known terms:
- ontology
- NGAC
- neatline
- NSDI
- conflation
- kriging
- DRG
- blob
- cartogram
- GML
- map algebra
My first thought is that these terms are simply not used much in 2011 and perhaps some harken back to an earlier time. I noted blob falls into that category and so, perhaps, does conflation. Does map algebra also fall into the “dated” category? I know it’s still widely done in practice as an analytical method but perhaps a broader term like raster analysis has overshadowed it? Editor in Chief Joe Francica suggests it’s less-known because the functionality has been buried as just another feature/function within image processing software and nobody has ever had to “group” these functions under the umbrella term as “map algebra.” He believes this goes to the heart of understanding the nomenclature of advanced spatial analysis, which few study anymore. Francica suggests [added 4/13/11 to clarify per comment below] analysts are “button pushers” without understanding the principals. If you told someone to “do the math” for raster data, he argues, that person wouldn’t know what you were talking about.
Kriging, while part of many GIS software packages, is probably not an introductory analysis tool, either.
I am surprised to see neatline and cartogram in this group since I expected core cartography terms to score higher. I’m saddened NSDI, national spatial data infrastructure, which I do not believe is a U.S. centric term, is not better known. I feel the same way about GML, geography markup language, an open standard for encoding geodata. Perhaps the lack of understanding of those key terms helps explain why it’s so difficult to build an NSDI on open standards in some parts of the world… like the United States? Francica points out this limitation may be an artifact of the lack of standards in the education of GIS practitioners. He points to the Geospatial Technology Competency Model (GTCM) as one tool to help in standardizing what GIS students learn.
Insights
What, if anything, can we take away from these results and analysis? I’ll toss out some observations, but am quick to point out these are qualitative, not rigorously science-based:
- We are less familiar with terms related to U.S. federal efforts.
- We are more familiar with terms considered “hot” in today’s GIS and technology arena.
- We are less familiar with terms widely used 10 or more years ago.
- We are more familiar with terms used in the manuals and day-to-day workflows of the software we use.
- We know some, but not other, terms connected to cartography.
- We are less familiar with terms related to standards and scientific analysis.