Computers, Society, and Nature

Much like Mark (2003) , Kuhn (2012) seeks to create a comprehensive list of core concepts in GIScience. Kuhn emphasizes the multi-disciplinarity of GIScience, and its importance in the growth GIScience. In general, I think that multi-disciplinarity is beneficial to any field, as different perspectives can provide fresh outlooks. Kuhn’s list of 10 core concepts is approachable for researchers in many disciplines, which can help promote cross-disciplinary GIScience research.

The core concepts are all relatively basic, but Kuhn’s more philosophical approach to them is really interesting. I found the discussion of location and accuracy particularly thought-provoking. Kuhn states that nothing has a true location, as location is based on relativity and context. While I immediately agreed that the understanding of a location is based on context, it took me a while to wrap my head around the fact that a theoretically unmoving object’s location would necessarily be relative to something else in order to establish its location (ie. I can’t be ‘here’ unless there is a ‘there’). I haven’t ever considered location as a dualism, but Kuhn has opened up my mind to the notion.

In the discussion on accuracy, Kuhn suggests that one aspect of accuracy depends on regularity in repeated measurements, but goes on to say that measurements must be understood as a random process. I will readily agree that units of measurement can be random, but shouldn’t the outcome of a measurement be far from random? Either I haven’t thought t enough about it, or a more in-depth explanation of this would be helpful.

In Geographic Information Science: Defining the field, Mark (2003) presents the “intellectual scope” of GIScience, and seeks to precisely define the science in a way that Goodchild (1992) did not do. He does this by suggesting that GIScience is a multi-disciplinary branch of information science, comparable to computer science. Mark clearly lays out the basic tenets of GIScience, and, in my opinion, successfully presents Geographic Information Science as a discipline that reaches far beyond GISystems and their applications. While reading the article, I often thought about the shortcomings in my GIS education during my undergraduate degree. GIS was presented almost exclusively as GISystems, and this article helped to provide me with a  base understanding of GIScience, its scope and its importance.

One thing that stood out to me, is that Mark suggests that GIScience became a truly academic field when the National Scientific Fund of the US began funding GIScience research. I find the use of agency funding as a source of science legitimacy incredibly interesting. On the one hand, funding is a crucial component of science and academia: without any funding, scientific research cannot be done. On the other hand, the fact that Mark’s understanding of the basic components of GIScience rests partially on successful funding proposals seems troubling. If government funding agencies have the power to define the scope and content of the science, is the science moving forward freely?

While this is a philosophical and ethical question that speaks more to our society than GIScience itself, given the personal privacy concerns tied to GIScience that we discussed in class, I feel somewhat perturbed by this. If government funding is pushing forward the scope and content of GIScience, how will citizen and consumer rights be protected?

I enjoyed Goodchild’s article as it served more as an almost nostalgic reflection on GIS’s progress over the years since its beginning. Goodchild’s debate on whether GIS is a tool or science quite interesting as well, as he states chronologically in the 80’s the focus on measuring error, shifted to a focus on uncertainty in GIS applications in the 90’s. This to me reflects how GIS, being so linked to computers software so early on, can easily be confused to BE the computer/software versus the science/scientific reasoning that goes into setting the parameters and finding appropriate uses for it in the realm of science. Personally, in working with GIS I find this reasoning behind which stats technique, or projection, etc. to be more and more of a scientific judgement, which cannot be made by a computer with problems of scale and the Modifiable Areal Unit Problem, to name a few. Goodchild hints at GIS being multidisciplinary in itself in contrasting geodetic science, cartography being an artistic science, and photogrammetry being largely engineering/problem solving. I feel this too leaves lots of room for people not familiar with GIS to focus on one, while ignoring the other applications of it and the fact that GIS today is often used as an umbrella term encompassing all of these very multidisciplinary areas into one subject.

As for Goodchild’s predictions for the GI-future, I agree with him in that neogeography and VGI will be the ‘future’ of GIS, highlighting the increased use of user-generated information and being more open to the general public (either passively or actively). I feel it’s this increased participation in GIS over the years, from a very non-user friendly interface, to being incorporated into so many mobile apps that makes the future for GIS quite bright, even if it doesn’t particularly hold to it’s ‘hard science’ background/early aspirations. I find his comments on ‘knowing where everything is, all of the time’ eerily current with the privacy concerns we brought up in lecture regarding never being able to shut off our phones, and although we have not fully reached this statement (though we are getting close), I could see this intensifying in the near future, as well as possibly coupling this with biometric data such as step count, heartrate, and even health from increased prevalence in wearable technology. Geography (as defined by Goodchild as relating to the earth or close to it) is also challenged by new uses of GIS, such as in it’s neuroscience applications (where earth’s topology is replaced with that of a brain) or modern AR, where the space could simply mean a table or a sandbox. Whether or not these are things to look forward to will be interesting to debate, and I can only wonder what the next 20 year report on GIS will look like.

-MercatorGator ?

This paper by Goodchild (2010) is an assessment of the (then) current state of the field of GISciences. Goodchild talks about the beginnings of the field of GISciences, research accomplishments and current agendas, and future predictions for the field. I often struggle to develop a conceptual framework for GIS, so I was happy to see that this paper did just that. By defining the field as the intersection between computers, society, and humans, I feel that I have a much clearer understanding of what GIScience actually is and the disciplines that it was born out of. Although, I feel that it’s worth noting that this conceptual framework doesn’t explicitly mention anything spatial…

This paper left me thinking about where GIScience fits within existing bodies of geographic thought. Goodchild’s many references to Tobler reminded me of geography’s “quantitative revolution” in the mid 50s. It seems to me that GIScience as it is today is only possible because of previous efforts to develop the field of spatial science, which is based on rigorous statistical techniques and scientific ways of theorizing. I was then thinking more broadly about theoretical understandings of space, and discussions of absolute vs. relative space. Yes, dimensions of space can be measured, but space can also be experienced, tied to significant symbolic meaning, and transformed by the perspective of the individual. GIScience fits very well within absolute theories of space, but how can it be adapted to answer questions about relative space?

• janejacobs

This article uses an interesting combination of quantitative and qualitative methods to shed light on decision-making among crowdworkers. The quantitative data demonstrated strong correlations between willingness to perform tasks and socio-economic status of the destination, while the qualitative data provided rather direct responses that implied causality. As far as the position of crowdsourcing on the tool-science spectrum, I would place it firmly on the tool side, because it’s applications are so purely commercial, and the use of the technology doesn’t contribute in itself to the furthering of geographic knowledge. This study’s focus on decision-making reminds me of my proposed masters’ research, which involves a discrete choice experiment. Choice experiments identify several variables that are of importance to interviewees in making a certain decision. The variables are then combined at different values in order to make several scenarios to present to the interviewee, after which they are asked for their preference. The interviewer can then infer which of the variables was the most important to the decision. Applying such a method could be interesting for a study like this, because several attributes of the destination neighborhoods are distinct but interrelated, e.g. socio-economic status, crime and race. The qualitative results implied that crime was an attribute about which respondents were vary open in citing as a decision driver. By contrast, the extent to which socio-economic status and race are decision drivers would be quite difficult because many people would feel ashamed to say so openly. In this case a choice experiment might not get around this problem, though choosing neighborhoods solely on the basis of race and asking whether the person would be willing to serve that neighborhood could be a viable method. Answering these questions would have important implications for the ethical value of the sharing economy.

-Yojo

There is also a notable difference in the relationship between today’s two topics and GIScience as a discipline. While issues of scale are more clearly within GIScience, the sharing economy is one of those topics–along with, say, drones–where what’s most pertinent to discuss is how GIScience technologies (GPS, in this case) are employed, and what their wide-ranging effects on society might be. In these cases, I think a valid question is, what can GIScientists contribute to a conversation in the social sciences and humanities to further our understanding of these new technologies?

There is evidence of a certain conceit around the “sharing economy.” As Isaac argues, uber wouldn’t exist the same way in a better job market, and there appears to be a continual effort to reduce the proportion of profits going to labour–epitomised by the plan to eliminate the drivers. When we ponder these aspects of a GIScience-potentiated technology like uber, are we still “doing” GIScience the same way as when we talk about issues of scale? I’d argue that even if we are not, in a strict sense, that we should broaden our definition of what doing science is. Coming to the end of the semester, I’m increasingly convinced that scientists ought to be better versed in methods of critiquing and analyzing the influence of technologies on society, and that this sort of thinking should be incorporated into various scientific disciplines.

In this article, the authors discuss the problems associated with re-scaling data and possible tools for addressing these problems. Re-scaling is required in order to compare data sets that are collected at different scales. I find the article extremely dense and challenging, being very heavy on statistical theory, and the examples provided to give context are themselves quite hard to understand. The article did give importance to several topics that are also important in the study of uncertainty, namely the modifiable areal unit problem (MAUP) and spatial autocorrelation. It is important to understand heterogeneity at scales that are finer than the scale of the sampling. I wonder, however (and the authors may have answered this question in language that I could not understand), how one incorporates heterogeneity at larger scales when scaling up. While I came to understand the MAUP as a product of the process of aggregating small-scale data to a larger scale and masking heterogeneity in the process, I suppose that it could be equally described as a process of dividing large-scale data to a smaller scale, except that heterogeneity must be interpolated when going from a large to a small scale. Furthermore, though interpolation, a crucial tool of re-scaling, was not prominent in my own review of literature, it is relevant to the topic of uncertainty because it involves creating data were no actual measurements were taken, so that the uncertainty is basically absolute. I’m actually not sure if interpolation can be approached from a position of error, vagueness or ambiguity. I suppose that error would be applicable because the interpolated value could be cross-referenced by samples from the field.

• Yojo

In Marceau’s piece, the issue of scale is discussed at length (no pun intended), and raises many good points. Scale and complexity truly go hand in hand, as complex systems can be invariant to scale (fractal characteristics) – a strange but intriguing phenomena.

While the two topics are inherently linked, the issue of scale comes up much more often, as it is very visible (scales at the bottom of maps) and important (“zooming” in and out on Google Maps, for example, to see the “bigger picture”). That being said, just because map users know what scale is, does not mean that they understand how it changes the information represented on the static or dynamic interface.

Marceau stresses the important of recognizing the Modifiable Areal Unit Problem (MAUP) – an important statistical error born from the aggregation of data over (typically) large swaths of area – and correcting any spatial analysis that may be affected by it accordingly. I do not pretend to fully understand the geostatistical implications of the MAUP, but I do agree that it is indeed a problem, and am happy that someone who understands the problem mathematically is working hard to find statistical solutions for it.

It is interesting to think about how the increasing use of dynamic interfaces such as mobile applications is changing how we reconcile issues of scale. As we can “zoom” in and out so easily, developers of future maps will have to generate many tiles to accommodate the users’ requests of displaying information at various scales. And to generate these tiles, we will have to really work through the MAUP, and by “we”, I mean not just “map makers”, but map users and map builders too. Will we have to include warnings at the bottom of these dynamic maps that “objects on map may not be closer than as they appear”?

-ClaireM

I found Marceau’s article to be a clear and easy-to-understand explanation of spatial scale, different frameworks of space and scale, and problems to do with spatial scale. I realized that I had really only thought of space, and therefore spatial scales, in the absolute sense, and I am looking forward to understanding the relative sense more fully.

This article made me think of discussions of how to incorporate qualitative data and methods in critical GIS. How would one go about using qualitative data while being cognisant of the problems presented here with spatial scales? From what I could find, there has not been much explicit discussion of spatial scales in qualitative GIS. However, I did find an interesting piece by Knigge and Cope (2009) in Qualitative GIS that relates the two topics. They use interviews and conversations to explore residents’ ideas of the vacancies on a rundown commercial street in Buffalo NY. They argue that the social production of scale is dependent on multiple processes (such as economic exchanges) and discursive practices, such as the imagining of “the city” or “the neighborhood.” They indicate that the scale at which data was collected revealed different interpretations of vacancy, which often conflict one another. However, one question that this paper brought up for me was the fact that the authors were examining this issue “through the lens of scale” – so does this mean that scale is just another lens through which problems can be explored, and therefore a lens that can be disregarded when it isn’t relevant? To what extent is scale a fundamental geographical issue that is necessary to address – or is it only relevant when it is causing these problems that Marceau talks about?

I may be in a bit over my head in trying to relate the very complex and nuanced topics of qualitative GIS and spatial scales, but I think there is definitely room for more research on the intersection of these subjects.

~ denasaur

Knigge, L., & Cope, M. (2009). Grounded visualization and scale: A recursive analysis of community spaces. Qualitative GIS. A mixed methods approach, 95-114.

This article emphasizes the importance of spatial scale in research and defines important concepts like space and scaling. Written in 1999, this article continues to be relevant to problems of scale presented by new technologies like drones. Marceau states “nor is a single scale sufficient to investigate phenomena that are inherently hierarchical in space.” She explains that doing this can severely jeopardize your research by hiding the modifiable areal unit problem. One of the important contributions of remote sensing, and more recently programmable drones, is the ability to rapidly collect data on phenomenon at multiple scales. In terms of mitigating the MAUP, the use of a drone to collect imagery could allow the researcher to perform a more robust sensitivity analysis.

I found the discussion on the difference between relative space and absolute space. The author writes that scale is the window in which we view the world, and that scales within relative space are more difficult to define than scales in absolute space, for example in remote sensing. As we move towards more advanced remote sensing using autonomous drones, I wonder how these concepts of space are programmed into AI. For example, traditional remote sensing uses GPS based imagery that is georeferenced in absolute space. But research is moving towards drones that can navigate absent of GPS coordinates, using computer vision to extract features from the landscape. This way, the drone can navigate around obstacles with only references to relative distance based on velocity and no computation of absolute space. Defining scale in such studies becomes difficult when the lines between absolute and relative space are blurred.

~anontarian~

This article uses the example of Uber to explicate the downsides of the so-called sharing economy . The author argues that Uber is another step towards the new neoliberal economy where employees have no job security or benefits. A depressed job market creates a steady supply of drivers willing to work and GIS technology enables the service to function. Their website says “We’re bringing Uber to every major city in the world.” If you’re a taxi driver, the situation looks grim. However if you happen to be an experienced GIS analyst, Uber will offer you a 401k plan, gym membership, full health benefits, and paid vacations. GIS-enabled sharing economy technologies are said to be disruptive in the name of efficiency and a better consumer experience; but from the comparison of benefits between the tech community and the average worker, it is clear who is really being disrupted. The genius of Uber framing itself as a technology company rather than as a taxi service is not just a loophole to avoid regulation. Uber really is a technology company, using its commission from drivers to create ever better geospatial infrastructure. When driverless cars put the Uber drivers out of work, Uber is still well positioned to compete as a transportation and logistics firm.

Get educated folks, the end is near:
Uber Jobs: https://www.uber.com/jobs/57019

-ANontarian