Computers, Society, and Nature

“Towards an integrated science of movement” by Miller et. al lays out the advances that have been made in the understanding of mobility and movement as a whole given the growth of location-aware technologies, which have provided much more accessible data acquisition. They are interested in synergizing the components of animal movement ecology and human mobility science to promote a science of movement.

In regards to mobile entities that are defined as “individually identifiable things that can change their location frequently with respect to time”, are there specific definitions that clearly define what “frequently in time” means? Examples have been made with birds or humans, but would trees or continental masses be considered mobiles entities as well?

It would be interesting to assess the impact of tracking location on the observations, in other words if tracking can affect the decisions made by whoever or whatever is being tracked. For example, a human who knows they are being tracked might change their trajectory solely based on the fact they do not want to potentially compromise sensitive areas or locations they visit, while an animal could behave differently if the technology used to track its movement make it more visible to predators. There is an ethical dilemma in tracking a human being without their consent, but it must be acknowledged that tracking does come with some consequences in terms of results differing from reality.

Analyzing big data is an obstacle across GIS, and movement is no exception. Cutting out potentially unnecessary components of the data in order to reduce the dataset  is one way of addressing this challenge. In Paul et al.’s piece they look at how much cutting down on datasets’ time windows may affect the end distribution.

Specifically, they examine the effects of changing the spatio-temporal scale of five different movement datasets, revealing which metrics are best to compare human relationships to movement across datasets. The findings of the study, which examines GPS data from undergraduate students, graduate students, schoolchildren, and working people, reveal that changing temporal sampling periods does affect the distributions across datasets, but the extent of this change is reliant on the dataset.

After reading this piece, I would like to understand more about how researchers studying movement address privacy. I’m sure having enormous datasets of anonymized data addresses part of this issue; however, I’m sure different government agencies, organizations, corporations, etc. collecting this data have different standards regarding the importance of privacy. How strictly enforced are data privacy laws (looking at movement data specifically)? 

This article by Fisher et. Al clearly lays out the components and concepts that are part of spatial data uncertainty and explain solutions that have been proposed to counteract their potential consequences on data analysis and interpretation. A better understanding of what uncertainty really is helped me realize that an overwhelming majority of geographical concepts are poorly defined objects, either being vague or ambiguous.

One solution for reducing the effects of discord ambiguity, although maybe not realistic but very practical, would be to create a global lexicon that stipulates how certain statistics need to be calculated and defines concepts on a global scale. This would allow for easier comparisons between regions currently using different approaches and would uniformize the process. However, it is important to note that this could not be applied to every statistical measurement, definition or observations made given the fact there could be biases against certain regions. An example could be that a road is conceptualized differently in one part of the world when compared to another.

On the topic of data quality, the advent of geolocational technologies has propelled geospatial data to the forefront of organizations and businesses aiming to profit from their use. Without trying to be too cynical, wouldn’t private organizations have an incentive to manipulate the data quality at the detriment of others in order to benefit themselves? This is where Volunteered Geographic Information (VGI), an example being OpenStreetMap, comes into play as to balance the playing field, in this case being Google Maps.