Is GIS a Wampeter?

A rite of passage for faculty members new to Penn State's Department of Geography is its longstanding Friday afternoon "Coffee Hour" series (Lewis and Zelinsky 1987). In December 1989, at the end of my first year at Penn State, I presented my first Coffee Hour talk, entitled "Is the Map a Wampeter?" The whimsical title referred to a word coined by Kurt Vonnegut, whose darkly comic novels meant so much to me and many others who came of age in the late 1960s and early '70s. A wampeter is "an object around which the lives of otherwise unrelated people revolve" (Vonnegut 1974, p. xv). The question my Coffee Hour talk considered was "to what extent does geographers' common interest in maps and mapping lend coherence to our diverse discipline?" Seventeen years later, this opportunity to write a guest editorial for Transactions in GIS prompts a similar question: Is GIS a technology around which the lives of otherwise unrelated people revolve, or has a "GIS profession" begun to coalesce? My responsibilities have evolved considerably since I arrived in State College as a lecturer and staff cartographer in 1989. Now I direct an institute for e-learning whose portfolio includes the Department of Geography's professional masters degree and certificate programs in GIS. As these online professional programs have increased in popularity (they now serve almost double the number of degree-seekers in the Department's academic programs on campus), my interest in the professionalization of the field has grown. In addition, my service as chair of the University Consortium for Geographic Information Science's (UCGIS) education committee, and as board member of the GIS Certification Institute, have also involved a quest for coherence in what has been called the "geographic information science and technology" (GIS&T) field (Marble et al. 2003). GIS&T is a remarkably broad and integrative conception of the geospatial enterprise. As illustrated in Figure 1, it encompasses three subdomains: Geographic Information Science, the multidisciplinary research enterprise that addresses the nature of geographic information and the application of geospatial technologies to basic scientific questions; Geospatial Technology, the specialized set of information technologies that support data acquisition, data storage and manipulation, data analysis, and visualization of georeferenced data; and Applications of GIScience and Technology, the increasingly diverse uses of geospatial technology in government, industry, and academia. The number and variety of fields that apply geospatial technologies is suggested in Figure 1 by the stack of "various application domains." UCGIS' Model Curricula initiative is based upon the proposition that the diverse practitioners and activities that comprise the GIS&T field do share a common, distinctive, and rapidly evolving, intellectual heritage. The three sub-domains comprising the GIS&T domain, in relation to allied fields. Two-way relations that are half-dashed represent asymmetrical contributions between allied fields. © 2006 Association of American Geographers and University Consortium for Geographic Information Science. Used by permission. All rights reserved The Model Curricula initiative arose from a set of eight education challenges identified at the 1997 UCGIS Summer Assembly in Bar Harbor, Maine. One challenge concluded that "improving GIScience education requires the specification and assessment of curricula for a wide range of student constituencies" (Kemp and Wright 1997, p. 4). A Model Curricula Task Force, chaired by Duane Marble, was formed in 1998. In 2003 the Taskforce issued a "Strawman Report" that presented an ambitious vision of how higher education should prepare students for success in the variety of professions that rely upon geospatial technologies (Marble et al. 2003). Central to the Model Curricula vision is a comprehensive "body of knowledge" that specifies what current and aspiring geospatial professionals need to know and be able to do. Following over seven years of deliberations involving more than 70 contributors and reviewers, the Association of American Geographers published the first edition of the GIS&T Body of Knowledge (BoK 1/e) in 2006 (DiBiase et al. 2006). Like the bodies of knowledge included in recent Computing Curricula (ACM/IEEE 2001), BoK 1/e represents the GIS&T knowledge domain as a hierarchical list of knowledge areas, units, topics, and educational objectives. The ten knowledge areas and 73 units that comprise BoK 1/e are shown in Table 1. Twenty-five "core" units (those in which all graduates of a degree or certificate program should be able to demonstrate some level of mastery) are shown in bold type. Not shown are the 329 topics that make up the units, or the 1,660 education objectives by which topics are defined. One of the most extensive of the ten knowledge areas in the BoK 1/e is "Analytical Methods." Twelve units, three of which are core units, comprise knowledge area AM (see Table 1). Fifty-nine topics, defined in terms of 281 educational objectives, comprise the twelve units. In many cases, objectives span the six "cognitive levels" and first three "knowledge types" identified in the Taxonomy of Education Objectives (Anderson and Krathwohl 2001). Also provided at the end of the knowledge area are references to 34 "key readings." An example core unit – AM4 – appears in Table 2. BoK 1/e will be criticized on various grounds, including which topics and objectives were included or left out, how included topics were parsed into knowledge areas, the extent to which educational objectives span the range of cognitive levels and knowledge types, and how cross-cutting themes are distributed among knowledge areas. Some will question the motives of any attempt to define the content of the field, since any such orthodoxy will necessarily exclude some knowledge, activities, and people. BoK 1/e does represent a claim that a GIS&T field exists, and that its diverse practitioners share a corpus of theory and knowledge. It is a claim that GIS is not a wampeter. And as the first formal, community-authored attempt to take stock of the GIS&T domain in a truly comprehensive way, BoK 1/e embodies one of Pugh's (1989) six attributes of a profession. Other attributes include cast of mind (i.e. self awareness); a shared social ideal; a formal professional organization; a code of ethics; and a "hall of fame" that recognizes leaders in the field. For Vonnegut, the prototypical wampeter is the Holy Grail – the enchanted tableware used by Jesus at the Last Supper. According to legend, Arthurian knights launched crusades to recover the Grail after it was lost by unworthy stewards. Eventually, the Grail became a metaphor for personal quests for redemption and spiritual perfection. In my 1989 Coffee Hour talk, I suggested that scholarship in cartography had been guided by a kind of Grail quest – the quest for the perfect map. The perfect map, I suggested, is accurate, unambiguous, objective, freely and universally available, and beautiful. Perfection was, and still is, elusive on every count except. Except, in some cases, the last. In 2007, is it possible to conceive of a "perfect GIS"? Such a system would include complete, up-to-date, and flawless data at any scale; the ability to model and predict complex phenomena; and an intuitive interface. Software code would be free and open, and universally accessible. And the system's default maps would be beautiful! As the ongoing research and development reported in this and other scholarly journals demonstrates, however, perfection in geographic information systems remains as elusive as the perfect map. Absent a perfect GIS, society needs GIS professionals who are knowledgeable and skillful enough to exploit the analytical capabilities of their systems, yet conscientious about ensuring that decision makers understand the limitations of GIS analyses. There may or may not be a GIS or GIS&T profession that fulfills all six of Pugh's criteria; the continuing lack of consensus on this point is documented in Goodchild and Kemp (1992), Obermeyer (1993), Wayne (2003), and GITA (2005). There certainly are GIS professionals, however, or at least people who wish to be recognized as such. Evidence includes the nearly 1,300 individuals who paid $250 each to be certified as "Geographic Information System Professionals" by the GIS Certification Institute between January 2004 and September 2006. Closer to home, over 400 new students enroll in Penn State's professional certificate and masters degree programs each year, and many others are turned away because classes are full. And the projected demand for "geospatial technology professionals" (whatever that means) is such that the U.S. Department of Labor has designated "geospatial technology" as a high-growth technology field (Sietzen 2004). What does it mean, then, to be professional? I suggest that it means being both competent in one's field and reflective about the ethical implications of one's work. Two organizations – GISCI and ASPRS – have promulgated codes of ethics to guide GIS professionals in matters of conscience. Job ads, and the professional certification programs of these two bodies, rely on the combination of education and experience as a surrogate measure of competence. A "GIS&T education infrastructure" has emerged to serve the field's educational needs. This is the constellation of educational institutions, government agencies, professional associations, and private businesses that cultivate formal and informal learning with, and learning about, GIS&T. Formal educational programs span a lifetime of learning – from primary and secondary schools through colleges and universities to continuing professional development (DiBiase et al. 2006). Several commentators observe that the GIS&T education infrastructure is producing too few highly-qualified graduates. According to Mike Phoenix (2000, p. 13) "the shortfall in producing individuals with an advanced level of GIS education is around 3,000 to 4,000 [annually] in the U.S. alone." Worldwide, Phoenix (2004) estimates that only about 500 students are pursuing master's degrees in GIS or GIScience. Referring to both professional and academic programs, he stresses that "the few graduate programs now in place cannot meet the needs of the marketplace" (Phoenix 2000, p. 13). Considering that 85% of the approximately 400,000 masters degrees awarded annually in the U.S. "are what have come to be called practice-oriented, or professional, degrees" (LaPidus 2000, p. 6), the shortage of professional degree programs in GIS&T is interesting, and troubling. Although many certificate programs in GIS are available at the undergraduate and postbaccalaureate levels (Wikle 1999), those abbreviated programs of study are likely to address only a fraction of the topics outlined in BoK 1/e. What's the difference between professional and academic degree programs? My institution has established that the purpose of a professional program is to prepare students to effectively apply the current state of knowledge in a field, while academic programs are expected to prepare students to advance the state of knowledge in their fields. According to LaPidus (2000, p. 6), professional programs are more "specialized in terms of focus, applied in terms of content, … and depersonalized in that they often seek to shape students according to a predefined template of professional competencies." A team of workforce development specialists sponsored by NASA recently concluded that the "geospatial technology industry" needs workers who possess "a blend of technical, business, analytical, and interpersonal competencies" (Gaudet et al. 2003, p. 25). Professional programs may be more likely than academic programs to cultivate this wide range of competencies. If there is a clear need for professional bachelors and masters degree programs in GIS&T, what's stopping higher education institutions from developing and offering such programs? One answer is capacity. In an era of flat or declining public support for higher education, few institutions are able to add the staff and space needed to support new professional programs while continuing to serve existing academic programs. A deeper issue, however, is faculty qualifications and predilections. In an earlier TGIS editorial, Karen Kemp pointed out that: Academics are often reluctant to admit that they really know little about the world in which their students will work on a daily basis. We may build their tools or ferret out the foundations of the science on which they are built, but how many of us really understand what it is that a GIS professional does at work? (Kemp 2003, p. 162). Professional programs are often tailored to adult students, and may employ practitioners as faculty. The success of Penn State's professional masters degree program is due in large part to its reliance on non-tenure track, practitioner faculty – individuals with advanced, but often not terminal (PhD) degrees, who have some years of professional experience involving GIS, and whose duties are focused on adult education. Although most college and university geography departments seek diversity in terms of the gender, race, and ethnicity of faculty members, fewer are keen to diversify faculty roles and qualifications in response to the needs of both professional and academic clienteles. The organizations that do embrace this kind of diversity are the ones that will prosper in the realm of professional GIS&T education. GIS professionals themselves can help by lobbying for professional programs, by volunteering to serve on program advisory boards, and by sharing authentic case studies from their professional practice for instructional use. Portions of this editorial were adapted from the GIS&T Body of Knowledge, with kind permission of the publisher, the Association of American Geographers (AAG). The GIS&T Body of Knowledge is available at http://www.aag.org/bok © 2006 by the AAG and the University Consortium for Geographic Information Science (UCGIS); all rights reserved. Illustrations realized by Barbara Trapido-Lurie. Thanks to my co-editors, and to all contributors to the BoK 1/e. None of these is responsible for shortcomings of this editorial.