Tag Archives: Internet

“The Annotated World, building a more social map” presentation by Ed Parsons, a geospatial technologist from Google

Today I attended a seminar presentation by Ed Parsons, the geospatial technologist from Google, titled “The Annotated World, building a more social map”. Parsons presentation was hosted by the Geography department in University College Cork. Although geography is not my area of research, geomantics, GIS, and geospatial tagging are some of my research interests as a digital arts and humanities scholar. Parsons presentation demonstrates how disciplines of the humanities can be revived and transformed by incorporating the use of computational methodologies, theories and tools in traditional humanities’ research. Subsequently, the discipline, in this case, geography shifts from being the study of the man-land relationship to the study of the man-land-computer relationship. Therefore, ‘digital’ geography, if it could be called that, becomes not only relevant to those studying the features and phenomena of the land and the earth, but it becomes an interest for computer scientists and multinational corporations such as Google who are looking to mine geospatial data to enhance their digital services and products, and the experience of the interaction between humans and computers.

With the employment of digital technology in Geography, Google has transformed the discipline into a marketable, knowledge product. This is evident from the creation of highly successful digital services such as Google Maps (a web mapping service and technology) and Google Earth (a geographical information programme), which are responsible for the radical transformation of how users surf the internet and how the user interacts with space and time in a digital and physical sense. The transformation of the user’s interaction with space and time may not be an obvious point until one considers how a traditional cartographic map or a digital map are windows to the past because they offer a glimpse of geospatial history in 2D and 3D form. The Historical Imagery feature in Google Earth allows the user to look at satellite imagery of Earth and travel back in time 5, 10, 15 or 20 years ago. With this service, the user can observe how our environment is changing both naturally and unnaturally over time.

[Google Earth Tutorial video]

The main points of Parsons presentation are as follows:

(1): How Google is using geography and geospatial data to overcome the problems of human-computer interaction by helping the user to locate and increase the accuracy of their search results through GIS (Geographic Information System).

(2): How we are moving beyond the traditional cartographic map with annotated, digital maps.

(3): Google’s future developments.

(4): The death of the desktop computer.

According to Parsons, “Google is all about finding things,” but in order for Google to be able to find location-specific information, such as if one was searching for a plumber in Cork and not New York, then “Google must use geography to help answer these questions”. By identifying the whereabouts of users, Google can solve some of the problems of human-computer interaction by filtering information content and reducing data deluge or the amount of inappropriate or inaccurate search results to those that only match the user’s geospatial data. Parsons explains that “1 in 3 searches on Google are about places” and that every time one uses Google one is also using GIS because there are algorithms processing information behind the scenes. All this content is being brought together unstructured and the majority of this information is accessed from mobile devices, which raise the question about the status of the traditional desktop computer as a species on the verge of extinction.

Beyond Maps

Parsons claims we are moving beyond the traditional cartographic map, but the first steps of this movement involved the rather simplistic and limited digitisation of maps, which only provided static, digital images, thus failing to serve as a solution to the functional and conceptual limitations of printed maps. In other words, the printed map was first recreated in digital form before it became what it is today.

Parsons questions:
• “Are there other ways of encoding the cartography in maps?”
• “What is the underlying information [that can and should be expressed]?”

The maps we use today, however, are radically different from traditional cartographic maps or the aforementioned, first digital maps because they are not static and dimensionally restricted. Instead, today’s maps are dynamic and dimensionally unbound because they are capable of manipulation in 2D, 3D and 4D form. By 4D I mean ‘Time’ because today’s maps make the boundaries of temporality malleable in a digital sense because they provide the user a view of the past, present and future. Touring the planet Mars on Google Earth is one example of how today’s maps show us the future. As well as being capable of manipulating visuality and temporality in a geospatial context, Google Earth can transcend the biological limitation of the human body in the digital realm. One can visually explore the oceans on Google Earth 5.0 without having to get their feet wet, to use a submarine or having to hold one’s breath during submersion. Although one will most likely hold their breath when they become immersed in the spectacular oceanic imagery. Sea-ing is believing!

Not only are today’s maps dynamic, but their creation is entirely different to the solitary task of the cartographer’s unidirectional inscription of his/her knowledge with ink on to vellum or paper. Today’s maps, however, are interactive – their creation is a democratic process involving a bidirectional exchange of knowledge between consumer and producer. This collaborative collection of geospatial data is then harvested to write algorithms and to display information in pixelated form on digital devices.

The Annotated World

As Parsons remarks, “People define places” because place is a social construct. In order for one to gain a sense of what a place is like, one must build an idea about that place and what it means for people. “Cork needs people to describe it,” Parsons explains, and Google is drawing from the online well of collaborative knowledge by requesting user participation in reviewing and rating places to construct a shareable sense of place. Parsons reminds his audience of the often-quoted joke, “On the internet no body knows you’re a dog,” to emphasize the unreliability of anonymous online reviews and ratings. Google, however, ensures reliability by adding a social element to the reviewing process. For example, the places I intend to visit will already have been annotated by people I know, but “not everyone’s opinion is equal,” Parsons claims. For restaurant reviews, Google bought ZAGAT to add professional restaurateurs’s opinions to the amateur food critics reviews. “All this, Parsons claims, “is combined in annotating the world around us”.

In addition to annotation, today’s maps are personalised and customised, which increases intimacy in human-computer interactions because it creates a sense of what I call artificial familiarity between the user and their computer. Personalisation and customisation humanizes our digital devices and gives the false impression that it is conscious and aware of our needs and desires. One of the ways Google achieves artificial familiarity is by offering the user search result choices that are dependent upon temporality – times of the day. To use Parsons’s example, if one is searching Google for places to eat late on a Friday night then the choices provided will be based on one’s previous eating habits appropriate to that time of day. If one is searching for places to eat on a week day then Google will not offer one the fast food search results of a kebah or burger shop that one wanted when socialising on a Friday night. This is another example of human-computer interaction where the computer is endowed with human qualities. In this case, our customised and personalised digital devices demonstrate the ability to learn and remember from previous Internet search requests and selected results. When digital technology becomes too customised and personalised, however, Parsons claims it crosses the boundary called the “creepy line”.

The Future: Where is Google going with GIS?

One of the possible future developments of Google in the next 5 years or so that Parsons discusses is 3D maps. The advantage of these maps, in comparison to today’s maps, is that they are captured in single time in 2.5D. 3D maps of the future will give a photorealistic view of the world, but Parsons claims they are not useful for particular tasks because “the real world gets in the way of information”. In other words, the annotation is compromised by the heightened visuality in the maps. Therefore, Parsons considers 3D maps to be mere “eye candy”, which differs significantly from the enhanced functionality of the hybrid approach – a second potential future development of Google, which Parsons discusses.

One area of geography that is largely unmapped by Google is the great indoors. Parsons claims, “70 of our lives are spent indoors or inside”. Mapping this uncharted geographical frontier was challenged by the unavailability of Wi-Fi in public buildings such as shopping malls and retail stores, but it is also challenged by the task of gaining permission to map a private space. With the increased availability of free and customer restricted Wi-Fi access in retail stores, Google can use Wi-Fi to locate and track the movements of shoppers as they navigate the store, thus creating a consumer-created blueprint of the inside of buildings from the geospatial data transmitted to Google from user’s handheld digital devices. Tracking and mapping our whereabouts outdoors is one thing, but tracking and mapping were we go and what we do indoors takes a step too far in to the private zone. Although it takes the breech of privacy to a whole new level, one group of people that would greatly benefit from the invasion of our indoor lives would be those interested in data mining consumer-behaviour analysis. If used correctly, consumer-behaviour data could revolutionise how people shop and experience products within shops, which are often hampered by a general lack of accurate, observational knowledge about how customers navigate and interact with products in store, which is essential knowledge for sustaining and maintaining a healthy and happy retailer-consumer relationship.

Google Now

“The map of the future, Parsons explains, “is not a map” because maps are not good for representing geospatial data.

Google Now uses context driven information. The example Parsons uses is notifications of traffic conditions that are not provided by sensors tracking the movement of traffic on roads, but instead they come from users. Millions of users are moving around and Google is using their geospatial data to generate real-time information content that is not requested by the user, but the geospatial information is sent directly to them. “All information,” Parsons emphasizes “is geospatial and does not need a map”.

Death of the desktop computer

As I mentioned earlier, Parsons noted that the majority of Google search results are attained from handheld digital devices instead of desktop computers. Parsons claims, “we have come to the end of that metaphor”. The desktop metaphor is being replaced by the (biblical?) metaphor of the tablet – the tablet computer, such as Apple’s iPad. Unlike Moses’s tablet, however, today’s tablet computer is not designed to operate after being broken in two during a fit of rage.

The desktop computer is also at threat from an up-and-coming hands-free digital device that Parsons says might be available to purchase by this time next Christmas. Google Glass are augmented reality glasses that give a more direct interface with the user from their direction approximation with our eyes. Google Glass combines the virtual and the physical world, and the magic of this amalgamation can be found in the middle ground between the two. A product of that magic is ambient information (http://itlaw.wikia.com/wiki/Ambient_information).

I mentioned earlier about how in human-computer interaction the computer is created to be pseudo-human or quasi-human in its interactions with us. With Google Glass, however, it could be argued that the computer is not only being created to simulate the ‘human,’ but it is becoming a part of the human – an inseparable part, to be precise, because Parsons informs his audience that Google wants us to wear Google Glass 24/7.

Wearable computers are attractive in numerous ways, but if one’s location and behaviour is being tracked 24/7 because our devices will be attached to us all the time then are we any different from the micro chipped animals whose migrations patterns we observe?

When we embrace digital technology, we’re extending ourselves, transcending the limitations of our biological and mental selves, but at the expense of our privacy and our precious connection to reality – it’s a double-edged sword!

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Data Visualisation – Anaesthetic or Aesthetic?

In my last blog post I discussed “The State of the digital Humanities: a report and a critique” by Alan Liu. In his article, Liu argues that the digital humanities is missing what he calls “Data aesthetics” (27). The list, line or bar graphs and tag clouds, Liu contends, reflect “the near-total imaginative poverty of the field in crafting an aesthetics of data” (27). Despite Liu’s claims that there is more focus in the digital humanities on metadata than “the look-and-feel of data” (27), however, the “availability and democratization of data” has brought about a vast increase and popular demand for data and information visualization (Lang n.pag.).

Data visualization can be defined as “the use of computer-supported, interactive, visual representations of data to amplify cognition” (Card 1). Data visualisation, however, not only amplifies cognition, but it also helps one deal with the “Data glut (McCandless),” by reducing “the mental load” on the user (Cawthon & Moere 2). David McCandless maintains that using our eyes is one way of coping with information overload because sight is the fastest of our senses. Eyesight has the same bandwidth as a computer network. Using aesthetic visualisations, therefore, facilitates knowledge compression and increases the speed of knowledge digestion because it stimulates the visual cortex of the user’s brain. In a sense, data visualisations resemble works of art to infiltrate the mind of the user. As McCandless notes, visualisation creates a new language that alters our perceptions. The interactive component of data visualisation allows us to participate as “data detectives” in the search for hidden clues exposed by the visualisation process. As Jer Thorp observes, visualisation makes data human by putting it in a human context. When data is sewn into the fabric of the real world, it gains meaning and weight. Realising this fact alters our dialogue with the information we are attempting to transform into knowledge.

Having said that, however, data visualisation is not a fool proof endeavour because a picture is not worth a thousand words if the viewer cannot decipher it. Nick Cawthon and Andrew Vande Moere observe, “the notion of beauty is not a normative element”. Perception of aesthetics is highly subjective. Therefore, an object cannot be viewed in isolation to its social environment and socio-cultural context because of the cultural and cross-cultural differences in visual language interpretation. The pitfalls of visualisation, Bresciani and Epplen claim, are “due to the fact that the meaning of symbols and colours are not universal” (11). This is evident from the foreground-background preferential differences in art of Western and Asian cultures, the cultural differences in the meaning of the colours red and green, and the way some eastern countries display time in a right to left format. As well as the problem of cultural bias, both the data visualisation user and designer require visual literacy and previous knowledge and experience with interpreting graphical displays. The psychological and aesthetic restrictions of data visualisation can confuse the user if the inherent meaning of the visualisation is ambiguous, but ambiguity may have a positive effect by effectuating new insights through creative interpretation of graphical depictions of data.

Questions to consider:

(1): Is data visualisation an aesthetic or anaesthetic? Does it enliven or numb the mind of the user through the graphical representation of abstract data?

(2): Why is aesthetics an important factor in information visualisation?

(3): Is visualisation a technology, a science or art? Does it have aesthetic value or merely aesthetic pleasure? Does data visualisation dilute perceptions of art or broaden its conceptual boundaries? Does data visualisation have to be artistic to be effective?

(4):  Data visualisation could be considered as an interdisciplinary subject? Do you think it flawlessly integrate science, art and design?

(5): Does data visualisation distract from the main goal of knowledge transfer? Is it more than a form of decorating or aestheticising dull data?

Works Cited

Bresciani, Sabrina, and Martin J. Epplen. “The Risks of Visualisation: A Classification of Disadvantages Associated with Graphic Representations of Information”. 2008. PDF.

Card, S. K., et al. Readings in Information Visualisation: Using Vision to Think. San Diego: Academic P, 1999. Print.

Cawthon, Nick, and Andrew Vande Moere. “Qualities of Perceived Aesthetic in Data Visualisation”. 2007. PDF.

Lang, Alexander. “Aesthetics in Information Visualisation”. PDF.

Liu, Alan. “The State of the Digital Humanities: A Report and a Critique”. SAGE 11.4 (2012): 8-41. Web.

McCandless, David. “The Beauty of Data Visualisation”. TED. 2010. Presentation.

Thorpe, Jer. “Make Data more Human”. TED. 2011.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Define Digital Humanities?

‘Digital Humanities,’ in my opinion, is a wide-ranging umbrella term that is far too broad and inconstant (perpetually evolving) to give a coherent and accurate definition that does justice to the nature of the field (or discipline?) and encompasses its achievements. It is for this reason that anytime I am asked by an inquisitive stranger or acquaintance to define ‘DH’, my response usually is: “How much time do you have to spare?”

Having said that, some form of definition – although it may be exclusive and rather limited in its definitional scope – is better than no definition at all, but whether it is purposeful or useful to have a definition of ‘DH’ is questionable.

The first question that springs to mind when I attempt to formulate a definition of ‘DH’ is where to begin. Perhaps, the nomenclature itself would be a good starting point considering that it demarcates a transitional moment in the history of Humanities. The merging of technology with textual culture effectuated change with the reconceptualization of traditional practice, and the development of new fields of inquiry and new forms of learning, research and teaching, while preserving (temporarily?) the foundational aspects of the Humanities. Resultantly, knowledge has been opened up from the confines of the disciplinary boundaries of academic departments to an interdisciplinary educational ecosystem, and from the scholastic to the public domain via the digital realm of the Internet.

For me, ‘DH’ is obviously more than the mere intersection of Humanities and computer technologies, humanities through digital tools, or any other condescending overgeneralisation. In my opinion, I think we (Humans) are the definition of ‘DH’. “Computing is not about computers anymore. It is about living,” as Nicholas Negroponte claims. We (Humans) are the etymological root of the word ‘Humanities,’ and it is not only because of this fact that we are very much a part of the definition of ‘DH’. Humans define ‘DH’ through our interactions with computers, which in turn assist the interpretation and definition of who we are, as humans in this digital age.

The definition of ‘DH,’ in my opinion, is not in the eye of the beholder, but in the hands of the Digital Humanist, computational tool holder. ‘DH’ is what we create it to be, which subsequently (re)creates us.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Does digital technology fracture or facilitate the reintegration of disciplines?

Digital technology creates an intersection between disciplines because digital tools are what Susan Leigh Star calls “boundary objects” – they are “entities straddling the borders between groups,” as Christine L. Borgman explains.

“Digital libraries, Borgman argues, are a canonical form of boundary object because their content can be useful to multiple communities, allowing them to carry meaning across the borders of multiple groups” (153).

Digital tools and services, Borgman contends, ought to be “generalizable, scalable, and interoperable,” in order to generate a flexible e-Research infrastructure for multipurpose, trans-disciplinary and interdisciplinary usage (252).

Digital tools provide a platform for the opening up of scholarly communication between disciplines and William Pannapacker envisions a future of disciplinary integration. Pannapacker claims: “We are reaching a consensus about the future of our profession that will involve not just language and literature, but all of the humanities in partnership with technologists, scientists, and information professionals”. Pannapacker predicts “it will become increasingly difficult to say what the humanities disciplines represent, by themselves—and to target them for elimination—because we are enmeshed increasingly in the transformation of every discipline in higher education”.

Consequently, it is my opinion, as the Three Musketeers’ pledge, digital technology and tools allow knowledge to be “all for one, and one for all” disciplines.

Borgman, Christine. Scholarship in the Digital Age: Information, Infrastructure and the Internet. Cambridge: MIT P, 2007. Print.

Pannapacker, William. “An Emerging Consensus.The Chronicle of Higher Education.

Star, Susan, Griesemer, James (1989). “Institutional Ecology, ‘Translations’ and Boundary Objects: Amateurs and Professionals in Berkeley’s Museum of Vertebrate Zoology, 1907-39“. Social Studies of Science 19.3: 387–420.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine