Computers, Society, and Nature

Prior to reading Langran and Chrisman’s article, my understanding of temporal geographic information was limited to time lapse videos of static snapshots, which visually display change. After reading the article, however, I was able to better comprehend the importance of temporal models that enable direct comparisons of how objects are changing. Models such as these allow for a greater understanding of how change is occurring at specific times, whereas snapshots seem to merely illustrate the general notion of change.

Outdoor Addict questions how decisions are made as to what constitutes an event and I think this is a valid concern. In abiding by data storage limitations, for example, we may deem a change to be irrelevant and discard it. However, what if this change is considered to be important at a future date? Perhaps the idea of examining snapshots is still holding me back—certain technologies that allow changes to be constantly tracked might need to be considered to a greater degree. In thinking about this, a parallel may lie in a Geography 407 class discussion, where dialogue revolved around sensors designed to continually track animals in forests. In this case, every motion that is detected is recorded, while durations of no motion are not. Can anyone else think of similar examples?

From the motion detection example, yet another concern arises—there will inevitably be information that sensors cannot detect. In addition, relating to the lecture on scale, the issue is not merely about deciding what objects to include, as previously mentioned, but it is also about determining what level of detail is appropriate. In other words, there is such a thing as too much information. The easy way out would be to yet again rely on the “future technological advances will render this concern irrelevant,” argument, but due to the inescapability of uncertainty, I posit that context and judgement are two of the most important considerations.

Lastly, in answering CimateNYC’s question about the distinction between real time and database time with regards to streaming data, Madskiier_JWong states that information may be incorrect or incomplete and in need of updating at a future date. I would like to add to this that technical issues often arise when dealing with streaming of data. For example, glitches in communication systems or backlogs of data can result in differences between real time and database time. This type of information is valuable, however, as it enables insight into how systems can be better designed.

– jeremy

The user-centered design proposed by Haklay and Tobon (2003) is very close to my own topic of critical GIS in that both topics recognize humans are an important factor in how GIS will be used and valued. Thus, historical, cultural and social aspects play a crucial part in the adoption and success of GIS. I especially appreciate how the article highlighted the fact that the improvement of GIS requires an “iterative process” between the tool and users (society). To maximize the practical usefulness of GIS, researchers must keep in mind that a “good” computer program cannot be judged solely on the number or the complexity of the application but rather on how “usable, safe, and functional” (579) the application is to users.

It makes logical sense that HCI research picked up momentum in the 1980s when personal computers became more affordable. However, I wonder what are differences between humans-computer interaction and human-computer interaction. Or in other words, between how groups interact with technology and how individuals do. Perhaps, decisions that involve a group of people (often in PPGIS), users may tend to listen to the one person with the most “expertise” and disregard their own knowledge of the application. The workshops described in the paper involved a user, a facilitator, and a “chauffeur”. I wonder if people would have interacted differently with the application if they were allowed “free-play” on their own after a short demo of the basic tasks.

Furthermore, I think we should carefully consider what tradeoffs are involved between usability and functionality. By making an application more intuitive and easier to use, are we losing important functions that should be included despite its complexity? Ultimately, this judgment depends on the set of tasks intended to be carried out by the application. However, they are not always easy to predict. For the purpose of planning, shortest path analysis may be extremely insightful, although the results may be difficult to intercept given all the assumptions that goes into the analysis. Moreover, uncertainty will definitely be another tricky area to convey. Therefore, one challenge is to figure out which types of tools should be included in a GIS for naïve users so that the system is both not limiting and not overwhelming.

Finally, the article made be think about the potential backlash of some HCI research. For example, I can imagine that disadvantaged communities may not want the results from the workshops to be published due to, perhaps, the misguided belief that making the system easier to use is equivalent to “dumbing it down” and the negative social stigma that follows. Therefore, the decision for including an opting-in or an opting-out option in HCI research is a sensitive one since this option will have dramatic effects on the number of participants. Personally, I favor having the initial settings to automatically include users in the study because although most people probably want to help and improve on a system that they use, the hassle of opting-in is enough to deter most people to becoming participants. However, due to privacy issues mentioned by the author, the application should explicitly warn users of this option before they can start using the application.

Ally_Nash

Haklay and Tobon stress the need to design both software and hardware that is most convenient for an identified user group and their goals. Cases such as Braille displays for computers are clear examples of a positive, improved human-computer interface. However, when applied to a complex analytical field such as GIS, HCI studies run into the issue of defining the user group. There is no immediate common attribute shared amongst all GIS users unlike the condition of being blind for Braille-users.

The authors are instructive when they indicate that identified difficulties with GIS are more human-based rather than technology-based. The challenge with users of GIS is that experience with the software varies wildly, and some may be unaware that they are a part of a GIS analysis (as a source of information or doing it themselves). It is tempting to simplify displays to extend the potential audience of GIS, and we have seen this in many GeoWeb 2.0 apps and platforms such as Geocommons. Geocommons’ site boasts that users can “Easily create rich interactive visualizations to solve problems without any experience using traditional mapping tools”. Web 3.0 continues in this direction by offering semantic searches and increasingly mobile applications and devices. In the drive to eliminate the distance between computers and humans however, it becomes easier for others to manipulate and use naïve users as data sources. The increased digitization of our world has sparked debates about privacy and perceived privacy issues.

A question HCI studies could ask then is how to best segment GIS users into groups. Can the semantic intelligence of Web 3.0 be used to map common thought processes/links to better identify common goals? This understanding can be used to direct intermediate GIS users to resources that explain the basics of analytic functions, while underweighting papers that describe advanced processes (I frequently ran into advanced “Petri Net” algorithms when searching for the general history of temporal GIS). Are there alternatives to segmenting users by “skill”, which is a vague measure and largely prescriptive?  

-Madskiier_JWong

In the chapter 5 section by Goodchild, he goes over how a geolibrary would work conceptually. However, I think there are some problems already brought up in his description. I don’t think that having server catalogues where a server is in the possession of one (or multiple) specialised collection is necessarily a good idea. It doesn’t sound very efficient to me. The thing about data is that there isn’t equal demand for it everywhere. Some data is demanded more than others (more people search for the weather at a given location than say, the demographic composition of it). Therefore, if you restrict servers to having only a certain kind of specialised content, you would not be optimising your server loads. Instead, you’d get a bunch of servers that have very low traffic that isn’t cost effective, and maybe more servers for which traffic is so high that you need to expand them. I just got the feeling that the suggestion was to model the hardware layout after the layout of the data, but this is just going to be inefficient. It means more connections have to be made, maybe more servers (and the extra costs associated with that), and probably slow performance.

The kind of system being described doesn’t sound very future-proofed. It’s like our ontology discussion. What happens when a new category is created? What happens when a sub-category becomes more important or separated from it’s original category?

In the article on fuzzy spatial queries, I think it is easy to say that we need to be able to form a method of querying that can incorporate both defined an badly defined regions, but how could this ever be compatible with the chapter 5 description of how the library would actually work? I think that ill-defined regions are just that, and I think the best we can do is just use a search engine. If we try to structure it (as we must do if we want specialised servers), then we run into all sorts of problems I think you all have an idea about.

Finally, I’m getting the feeling that we are meant to start our queries in a geolibrary with a location (well defined or not). It seems that geolibraries would be tailored for a certain flow of querying (location > topic > sub-topic > person etc.). What if we don’t want to start our query with a location?

He’s talking about AltaVista….AltaVista people. Remember them? No? That’s because they’re dead.

-Peck

Chapter five by Goodchild provides a good overview to geolibraries, their importance and the components that goes into constructing them. He highlights the difference between geolibraries and traditional physical libraries. Namely, that geolibraries would be more compatible to deal with multimedia content, hold more local information and avoid the issue of duplication. I think today the separation between the digital and traditional library have become much less distinct. Online catalogues allow users to search many libraries at the same time and thus address the problem of duplication to some degree. As more books and other materials are digitized, a user no longer needs to go to the library to get the material he/she needs. I often download newspapers, articles, and magazines from the McGill library. Also we all frequently download maps and GIS data off the Internet especially since many cities have established open data portals. Thus, I argue, the key feature that will make geolibary special is not how comparable it is to physical libraries but how it allows users to discover various topics about a location of choice.

Further, I wonder why geolibraries have not become very popular since 1998 because this is the first time I have heard about them. Maybe it has to do with the 4th research question Goodchild asks: “What institutional structures would be needed by a geolibrary? What organizations might take a lead in its development?” I would also like to add, what kind of personnel training and organizational shift in the ways things gets done are required by current governmental structures to enable the adoption of geolibraries? There is definitely inertia within public office structures that is often difficult to overcome when introducing new technology. Finally, I would like to consider who should be responsible for the data and the limitations of the kind of data offered. I remember Peter telling me that one of the reasons why governments are hesitant to make data public was if the data contained a mistake about an area, who should be held responsibility for the damages? The one who collected the data? The one who entered the metadata? The geolibrary for providing bad data? Also, what kind of limitations should be set on the kind of data downloadable through a geolibrary. For example, restrictions should exist on data that are highly political such as health data or high-resolution environment data.

Ally_Nash

I find the topic of geo-libraries fascinating, particularly because of the incredible potential I think that they have in conveying ideas. It is a very powerful way of organizing information that allows for visual comparisons to be made. For example, related information that may be of use to a user may be more easily suggested or discovered, as information that is geographically related can displayed. The importance of geo-libraries in practice is also backed up, as the majority of map library users rank locational characteristics as the primary search key.

Since I’m a bit of a building/architecture nerd, I think the following is an interesting example of a geo-library:

http://skyscraperpage.com/cities/maps/

Perhaps this is a very simplified version of a geo-library, but it shows how an individual can search for information on proposed, under-construction, and/or competed buildings by selecting a geographic region. As one can see from the map, very useful visual information can be gained through pattern recognition. Clusterings of under construction buildings, for example, are easy to find.

While I think that this is an extremely useful tool, my example is one that may be much more straightforward than those discussed in Goodchild’s article. Most of the cities listed, for example, appear to have well-defined, uncontested boundaries (or perhaps they just appear that way?). Further, the uncertainty present in the information available for each building (location, type, developer, floor count) also seems to be relatively low.

Even though this example may be simplistic, I think that it points to the potential for geo-libraries to have a regional bias. By examining the list of cities available to search, for example, it is clear that this architecture website has a North American focus. However, perhaps technology will enable users to overcome this bias. An example provided by Goodchild is the use of geo-library ‘crawlers,’ which search the web looking for key terms based on geography. Despite their potential, these technologies also bring with them a variety of other problems. For instance, as mentioned by ClimateNYC, the issue of ontologies arises, where incorporating varying opinions and definitions proves to be troublesome.

Perhaps someone can answer this in their blog post, but I am uncertain as to the various forms geo-libraries can take. For example, since users can search for related businesses within a Google Maps map frame, would this be considered a type of geo-library?

– jeremy

In a geography of Asia class I took last semester, possible ways for individuals to engage in responsible tourism was discussed. Most often, much research was required in order to educating oneself on local conditions or current events. However, sifting through the biased perspectives of local media or governments can be a challenging activity. Especially when trying to understand how tourism impacts marginalized communities, understanding this can be an important factor. I think that a geo-library can facilitate learning in this case, where tourists can query and access information depending on what region they are in. The option of seeking out academic articles can enable people to gain scholarly perspectives that may more accurately represent local conditions.

As the article by Goodchild et al. mentions, however, an issue arises when footprints or terms of searches are ill-defined. With regards to tourism, it would seem that this problem would be magnified in remote areas. For example, as we know from our discussion on ontologies, defining objects is very challenging, especially when multiple languages are being considered. Further, less information is available for remote areas in general and so while geo-libraries may be extremely enabling in many aspects, they may also be limiting for areas not well-represented.

Perhaps an interesting comparison is likening a geo-library to a mental map. While the regions that I best understand will likely have the most amount of detail, it will not include important elements belonging to another individual’s mental map. As Goodchild et al. posit, objects with ill-defined terms that are not well-represented in a geo-library can also be of great significance to the lives of individuals at a local level. In other words, while a place name or building may not appear in a geo-library query (or my mental map), it may still be very relevant to many people. Determining how to incorporate under-represented features will be a challenging, but crucial issue in the development of geo-libraries.

– jeremy

Cyberinfrastructure’s post on the design of gazetteers sparked an investigative light. Taking a test run on the Alexandria Digital Library, a search for “Northern Canada” focuses in on the majority of Newfoundland. Immediately, as a GIS user with some appreciation of its underlying principles, I was struck by the lack of transparency and feedback on how the geolibrary decided on Northern Canada as referring to Newfoundland. I tried typing in the name of several towns in the territories, northern Ontario and BC to establish a search history that pointed towards my definition of northern Ontario (to see if the engine would contextualize future results based on previous ones). No such luck.

The service seems much more oriented towards the delivery of data and is more reticent towards acknowledging the uncertainty of the gazetteer and how people have different ideas of the extents of places. Cyberinfrastructure’s post alludes to a growing and increasing diversity of users to match the need for constantly updated gazetteers. As the number of users and their purposes for the data grow, it becomes increasingly important to explicitly discuss the (compound) uncertainty in the datasets provided.

Based on initial impressions, a rudimentary system can be developed that allows for more flexible, user-defined boundaries. If a user defines an area that overlaps two formalized regions in the gazetteer, the system could return datasets for both regions involved. More control needs to be given to the user to understand how things are geolocated within the ADL and to include more contextual fields that tailor searches to the user’s preference. These issues are clearly in line with Goodchild’s 7^th research concern on the cultural barriers inhibiting cognitive understanding of geolibraries. Presently, the ADL seems designed for planners that follow legal, administrative boundaries rather than for the general public user. Even simple contextual algorithms that Google uses in its search engine could be implemented to enhance the querying experience in ADL.

– Madskiier_JWong