Computers, Society, and Nature

Sieber and Sengupta’s paper on geospatial agents situates artificially intelligent agents within the context to GIScience. This spatial context results in the geographically oriented applications of the concept referred to as ‘Artificial Life Geospatial Agents’ (ALGA’s) and ‘Software Geospatial Agents’ (SGA’s). Effectively, these agents are technologies that allow GIScientist to be more efficient in their work by automating elements of the GIS process. These agents are limited by the information available to them; essentially the information available to them is the data that has been made public/accessible. The environments in which the agents are situated could lead to discrepancies among users depending on the information available to the user. in other terms, considering privacy, a user with extensive private data may have more applications and success with the agents than a user using only publicly available data.

I believe that the article handles the subject well within the domain of artificial intelligence. However, as someone with layman information regarding AI, I feel that more context regarding AI at the foundation might have helped to explain how these agents function; understanding the limitations, and the possibilities in addition to the applications might have helped clear up some of the ambiguity I had around the topics.

The article has encouraged me to explore these limitations and see what applications these agents might have for someone for a GIS user such as myself. It would be important to see how much information is needed on behalf of the user to properly take advantage of the geospatial agents.

I thought this review was an interesting contrast to the discourse presented in the Wright et al. (1997) article on GIS as a tool or science. There seems to have been a transition from GIScience arguing for its own existence to asserting its domain over established concepts.

At first, I was a little skeptical of the unique “geospatial” designation for agents used in GIScience. I was easily persuaded of the commonalities between the properties of intelligent agents described in AI research and those applied in GIScience. Perhaps too easily persuaded. I struggled with how geospatial agents could be distinguished from other intelligent agents–particular those that don’t explicitly operate in geographic space. The element of geographic space is more evident in the case of artificial life geospatial agents, but at a glance, a software geospatial agent used to locate and retrieve spatial data from the Internet might resemble any other intelligent agent used to scrape non-spatial data. Of course, handling any spatial information requires some understanding of topology, scale, spatial data structures, etc. that is inherent to GIScience. In fact, I would imagine many intelligent agents implemented outside the domain of GIScience could benefit from the nuance that GIScience is able to offer.

I’m convinced! Geospatial agents most definitely necessitate their own designation. Again I’m reminded of the plight of the neogeographer. The article demonstrates a clear need for GIScience considerations in what are sometimes careless applications of geospatial information in technology.

I thought Sengupta et al.’s article, “Automated Extraction of Movement Rationales for Building Agent-Based Models: Example of a Red Colobus Monkey Group” (forthcoming 2018), was incredibly interesting. “Automated Extraction” discusses the use of agent-based modeling (ABM) strategies in simulating red colobus monkey groups’ movements “across space and time and predict[ing] environmental outcomes” (2). Utilizing the knowledge of experts as input, the modelling hopes to augment “the expert’s interpretation” (2).

At the conclusion of the article, Sengupta et al. note the possibility of ABM’s eventual replacement of scientists’ “heuristic knowledge” (11). It is exciting that ABM is continuing in the theme of original excitement behind GIS (helping us identify patterns that are not easily discernible quantitatively). However, it is also incredibly worrying, as it has the possibility of growing larger than zoological research purposes.

Sengupta et al.’s model relies on human monitoring of the model to check for errors, and the model requires more information from human experts’ field observations to become better at modelling. If AI were to be introduced to the model, and the model learns and understands the patterns better than human experts have observed or can observe, could we reach a point where nothing is unpredictable?

Continuing with the animal theme, this information could be used to predict, for example, where a group could be at a given time and then used on wildlife reserves to organize tours with high success of tourists seeing animals, or help researchers with short time-tables to most effectively study the animals. However, if poachers were to access (possibly) highly accurate modelling, they could more accurately predict the location(s) of groups of animals on the reserve and become more effective hunters of protected species.

For applications using human populations, ABM could be used for humanitarian purposes, like finding the most ideal evacuation routes (and edit existing routes or add new ones) for natural disasters, for example. However, if the model learns extremely well and everything becomes predictable, what would stop nefarious actors from using this information on human populations to catastrophic degrees?

I found both of these articles really fascinating, and found it helpful to understand the theories and differences between the ALGAs and SGAs presented in Sengupta and Sieber (2007), especially in analyzing the ABM presented in Sengupta et al. (2018).

Sengupta et al. (2018) use field-recorded data on Red Colobus monkey location and movements in space and time, combined with other GIS data (land cover type, slope) to automate ABM movement rules in an ALGA. Sengupta and Sieber (2007) suggest that ALGAs began with studying the flocking behaviours of animals and birds. To me, this suggests that the effects of a study on animals can have broad-reaching effects, beyond the study of “movement ecology” (2).

Sengupta et al. (2018) suggest that the advancement of high-resolution tracking technologies have created an ‘“enormous volume” (2) of data (i.e. big data). In this study, the authors refer to big data derived from GPS tags on animals, but couldn’t this easily be expanded to the movement-tracking (big data-creating) devices we carry around with us all day? Is there justification for concern given the necessarily reductive and therefore inherently wrong nature of models? Are our movements as easy to predict as a Red Colobus Monkeys’?

Though I tend to be a bit a doomsday-ist and cynic, in this case, I think that though models may try to track behaviours and predict movements, both humans and animals have one things the models don’t have: free will. Though the models can incorporate and automate complex decision-making models, I think that humans and monkeys and various other animals sometimes make irrational decisions that models cannot predict. In fact, I think that there is a huge potential for error in these models, which neither article addresses.

Sieber’s article establishes the historical context for PPGIS, and explores a framework for evaluation based on themes found throughout the PPGIS literature. It’s an interesting point that the term “participation” itself suggests the need for some intermediary. If PPGIS is to be viewed as a decision making tool, I would imagine that the typical role of the intermediary is to facilitate the relationship between stakeholders and decision makers, perhaps by way of technical GIS expertise. When stakeholders are empowered by a “bottom-up process,” or their own decision-making power or technical expertise, does a PPGIS framework still hold? Is the ambiguity a problematic feature of PPGIS, or is it that it should be differentiated from PPGIS in some way?

I was really struck by the discussion about public participation as a “ladder of increasing involvement and influence in public policymaking.” Admittedly–maybe unsurprisingly as an MSE student–I’ve always accepted the idea that ascending the ladder of stakeholder engagement is the ultimate goal. Evidently, I suppose it’s important to consider the ways in which community control are realized. In the era of the geospatial web, it’s conceivable that community control through PPGIS would likely require some technical ability on behalf of the community members, perhaps access to the internet or a personal computer. Of course, challenges to the framework arise if the ability to meet these requirements varies between individuals. It’s clear one of the most critical aspects of the public participation GIS framework is the consideration for differential ability to participate among the public.

In this article, Sieber traces a history of PPGIS, engages with the existing literature to create a framework for PPGIS. I found lots of the discussion very interesting, but what I found most interesting was the discussion on the accessibility of data. As PPGIS involves those affected by decision-making in the process, accessibility to data is crucial. I have to admit I have not ever contemplated the definition of ‘access’, though the various definitions show the nuances in the understanding of ‘access’.

While reading the four competing ethics of data availability, I was struck by how each of these positions, and politics, could drastically alter the process of PPGIS. I am also struck with how fluid the boundaries between these ethics can be, and I think that most countries would employ a combination of these approaches to data availability.

An open government would facilitate PPGIS, while any of the other positions would hinder PPGIS to varying degrees. While I mainly agree with the open government position in terms of spatial data, I also understand why personal privacy is important, and can in fact be crucial to a healthy society. Likewise, in terms of national security, it could be important to protect the location of secure facilities. I do have fundamental issues with the fiscal responsibility position, and see this as the biggest hurdle to effective PPGIS (good old capitalism…). Putting a price on public data invariably grants access to resource rich organizations, solidifying a top-down framework of PPGIS. This touches on the notion of the inherent inequality in PPGIS, a subject that I think Sieber does a good job addressing throughout the article.

Sieber et al. (2016)’s discussion of Doing Public Participation on the Geospatial Web raises issue with accepting the standard GIS uses in governments as forms of public participation. While this public use seems benign and helpful at first glance, their analysis shows that it is not always this way, and the government-public relationship has remained fairly unchanged by the advent of the public use of the Geoweb.

The field of PPGIS will be particularly interesting as the “digital generation”, or those who have grown up or started from an early age in using the Internet or other virtual devices, ages. As more of this generation reaches voting age (usually when one becomes politically active/conscious) the ways in which government interacts with citizens will change irreversibly, and perhaps the demarcation between government and citizen will blur or mutate as well (as Sieber et al. denoted has only slightly occurred so far).

In addition, the aging of this “digital generation” may eliminate the digital inequality brought on by technological advances. I hesitate to say that it will eliminate this specific inequality, as the “broadening of access” excitedly brought on by technology clearly has not lessened the divides between urban-rural/socioeconomic/age, as Sieber et al. noted, but also between (dis)abilities, in accessing resources virtually. And will web-based tech ever lessen the divides?

Without intervention, lower-income communities or geographically isolated communities may not have access to the web due to lack of financial resources, lack of device availability (to buy or to rent), lack of a platform to connect to, or other such concerns. In addition, technology will always evolve and it will always be “new”, regardless of which platforms or tech or equipment on which the older generation grew up, and the older generation may continue to not have access to teachers or they may not care to learn how to use/benefit from new technologies. Finally, text-to-speech technologies have made advances in connecting sight/audio impaired communities to the Internet, but there remains a lack of access for those with motor skills impairments, for example, which hopefully will be solved with advancements in science. With these persisting issues with web connectivity, public participation through the Geoweb cannot be taken at face value and must be studied more thoroughly through an equity lens.

Goodchild (2010) provided an interesting summary of the innovations in GIScience since its first conception. I appreciate the discussion of what innovation looks like in a field that is so intensely interdisciplinary. I feel that Sara Fabrikant’s interpretation of “discoveries” in GIScience as “enabling the discovery of the world” also appropriately captures the interdisciplinary nature of GIScience research implications.

In the 7 years science the article was published, I wonder how successful we’ve been in addressing Goodchild’s fifth challenge of the decade: the challenge of education. He writes that in the first twenty years since the conception of GIScience there has been a transition in the approach to education toward GIScience literacy in the general public. Still, most of my technical GIS education has been with software otherwise locked behind costly licenses or affiliation with wealth institutions. I’d imagine this approach serves to reinforce a knowledge gap between those with access and those without.

Perhaps it’s because of a lack of suitable alternatives, and I can’t speak to the experience of students at other universities, but I think it’s at least indicative of some vestige of educating a professional elite in GIS education. Of course, universities have some duty to endow students with employable skills. I suppose another challenge in GIScience will be to provide students with a knowledge of industry standards without perpetuating a culture differentiated expertise based on resources.

In the paper “core concepts of spatial information for transdisciplinary research”, Kuhn(2012) demonstrates the importance of spatial information across disciplines and proposes ten core concepts for non-specialists to understand it. When reading it, we should question about whether Geographic Information Sciences (GIS) have to be transdisciplinary, and whether these ten core concepts are helpful for those kinds of research.

According to Kuhn(2012), spatial notions are the basis of transdisciplinary approaches. The example of the Amazon deforestation proves his statements which I agree with. In other words, we can regard spatial information as integrators to connect different disciplines. However, I would doubt that the ten core concepts are useful enough for people outside of GIS.

Non-specialists need to understand and apply these concepts in their practices. Therefore, the chosen concepts should appropriate for their use. Hence, I’m interested in the objects of the survey conducted by the author. Whom or what projects it refers to? What disciplines are involved? Does the information mainly provided by GIS-related practitioners or researchers? Answering these questions may help us know how useful the concepts could be for non-specialists. Besides, some of the descriptions of concepts are straightforward enough, while some are confused. For non-specialists, the isolate description of “field”, “object”, “meaning” or “value” is not able to make them well-understood. For example, “neighborhood” can be thought as a region, “field” also. “Field” can answer what is here, “object” also. “Meaning” and “value” sometimes may refer to the same thing. Non-specialists still have few concept of these “core concepts”. Unless, there is more discussion about the relationships among core concepts, the situation may be improved. That said, it is necessary to organize these core concepts so that they can be understood accurately. In conclusion, Kuhn notes that core concepts should be formalized into an ontology when use, but it remains to be asked whether we can have them more structured in ahead of the practice.

Initially I had some trouble appreciating the importance of the debate summarized by Wright et al. (1997). The discussants bring up some interesting points about the nature of science, but I’m curious about how productive the conversation can be beyond the need for some scientists to act as gatekeepers. Particularly in the sense that the science/non-science dichotomy, as noted by the authors, “presumes the superiority of one or another approaches to generate knowledge.” I was reminded of how similar discourse in the field of environmental science has been used to delegitimize traditional ecological knowledges–”similar” in how the discussants try to incorporate the concept of GIS into their own understanding of science, not necessarily in the context or implications of the discussion. I’m not so sure that science lends itself to such unambiguous ingroup/outgroup designations. I can, however, appreciate the very real challenge of securing research funding and by extension the need for asserting academic legitimacy.

I think how some discussants so boldly assert their opinions really demonstrates how strongly our ideas are shaped by our world view. For example, the argument that GIS is analogous to math and statistics and therefore not science is less compelling without the assumption that math is not a science. I was eager to see the authors draw the comparison between GIScience and computer science, which arose with the development of digital computing, a powerful tool for processing information. I imagine now few would question its place as a shining example of STEM-hood.

I found Goodchild’s argument refreshing, as it encompassed all the issues that I felt were present (but relatively left unsaid) during my time learning how to use GIS software. His discussion of the achievements of GIScientists (like Tobler’s First Law (“nearby things are more similar than distant things”) and Anselin’s spatial heterogeneity) really helped me to better understand the difference between GISoftware (tool) and GIScience. It became clearer to me that GIScience is its own domain, which helps it find applications in both theoretical and technical fields, including but not limited to computer science, economics, politics, and marketing/business.

He also acknowledged what I have found worrying about geography academia– the tendency of scholars to look solely to academia to bolster their argument.  In my experience, the papers that look solely to academia to make their claims papers end up rehashing the same arguments, not offering any new or terribly groundbreaking ideas. It is one thing to refute arguments and conduct new hands-on research into phenomena, but many that I have experienced as a liberal-arts undergraduate student rephrase and slightly elaborate on claims that have already been made, which Goodchild calls “problematic”.

Finally, Goodchild’s observations on the “future of GIScience” are very well thought out and many of his questions remain unanswered today. I was particularly interested in his discussion of user-volunteered data, particularly about the social strata of data volunteers as well as their motivations to collect and provide this data. I am particularly interested in participatory planning practices, and today, many of the ways of collecting public input involves some iteration of online surveys. Many towns– including my small (and fairly technologically challenged) hometown– conduct surveys which take longer than 30 minutes, requiring users to place points on online maps or drawing lines on online maps to show commonly taken routes, for example, in order to plan for a better use of space. Goodchild asks: who is actually volunteering this data, and why? I think some question that also need to be asked are: Who is not volunteering this data, and why? Does the lack of technology or time or interest (or, more likely, a combination of the three) dissuade everyone from volunteering this information? And how could those who do not volunteer this information (the elderly or others who do not own personal computers, people who work over 40 hours per week) provide this information in another way? Even 7 years after the publication of this reflective article, information about data volunteers is rarely studied (though sometimes in these surveys, this data is collected/volunteered) and therefore efforts to be more encompassing are often in name only.

I found the overall article creates more complexity than simplifying the proposed concepts. If the target of the article is to communicate the opportunity of interdisciplinary integration, I find the language used to express Geospatial Information’s primitives (and cores) too complex if not supported by pragmatic and clear examples.

In reading the article I question why the metadata concept is not part of the core discussion given the fundamental roles it plays in information and moreover in the geospatial domain. This concept could have been elaborated in the accuracy section.

The author provides an unequivocal interpretation of the core components in Geospatial Information. That said, he also let me think that in a society driven by a strong economical drive, measuring the economic value of Geospatial Information is complex and difficult to estimate.  So how can we estimate its impact on the budget planning of a project based activity?

This gap suggests me that the Geospatial Information could remain in the ‘Nice to have’ basket when planning project’s resource investment (which the primary focus in not Geospatial Information).

Lastly, because things change constantly, I am still debating where the concept of time sits in the Geospatial Information realm. Is it a core element or should it be considered an external factor?

– Giancarlo –