Epidemic modeling can be used to model not only the outbreak of diseases like SARS, but also the diffusion of an innovation. One interesting point that Watts raises is that “on a small-world network, the key to explosive growth of a disease is the shortcuts.” This leads me to ask what shortcuts exist between an innovator and a market. Is there a way to make the diffusion of an innovation more efficient?

We know from Everett Rogers that the diffusion of an innovation relies on its ability to flow from innovators to the early majority, and we know from Clayton Christensen that the leap between the early adopter and early majority groups is the most challenging period in this process. I aim to look at the people and organizations involved in that process to understand their characteristics and behavior. I believe we can identify actors that benefit the process and actors that lag or harm the process, and then identify the characteristics and behavior that produce a more rapid and/or impactful result in moving an innovation from the innovator to the market.

For example, in describing the impact of random rewirings on variously sized networks, Watts states that “the first five random rewirings reduce the average path length of the network by one-half, regardless of the size of the network.” This implies that the first five connectors in a network matter most in the connection between networks or nodes. This effect is amplified when those initial connectors are established (i.e. trustworthy) or well-connected individuals, in other words well-known in their field. In fact, as Watts describes, well-connected nodes have the luxury of attracting new links while poorly connected nodes remain poor. This is known as the Matthew Effect. So we might conclude that the connectivity (wealth of connections) of the initial five connectors determines the ego’s (or central node’s) fate. If you set up your network among loners, your network will stagnate; among the gregarious, your network will thrive.

“The Hacker Ethic” was written in 2001, following on the heels of Manuel Castells’s “The Information Age” that described the new information economy and network society. I wonder how Himanen would write the book any differently today. Would he break hackers into categories beyond “crackers” and “hackers,” perhaps distinguishing gamers or quantified self followers? Or are these all ultimately just hackers, with a common ethic? And is the line between crackers and hackers so thick today? Is Julian Assange a cracker or a hacker?

What does the hacker ethic look like across the globe? While it is possible that the Protestant ethic and its baggage was shipped with the personal computer, it is more likely that local mentality adapted the personal computer to its own rich heritage of ethics. How are hackers different in Seoul, in Lagos, in Sao Paolo?

What impact does the hacker ethic have on the relationship between hackers and their market and the actors involved in delivering an innovation to that market? Why do we see so few hackers succeed in the business world? What distinguishes those who do? Is their success determined by the alignment of their passion with their product and the market’s desire? In other words, does this spark of innovation happen only when the hacker’s passion matches the desire of the market? What is the relationship between hackers and entrepreneurs – are entrepreneurs in fact hackers? Or are the hacker’s desire to spread innovation freely and the entrepreneur’s desire to capitalize on it perpetually in conflict?

I plan to continue exploring the sociology of the hacker through literature and conversations with hackers and hacker experts.

I’ve mapped my LinkedIn network using inMaps, and have a thick curiosity about the characteristics of the people in my network that link across my different affinities. What is special about them? The vast majority of my network do not link beyond a single affinity. But a few do.

Some connections make perfect sense, such as a business professor from one school connecting with students and professors at another school. Connections across two affinities are not unusual either. The people I connect with are likely to have more than one thing in common with me, so they naturally know my connections in those other circles as well. But there are a few that have multiple common groups with me. This makes me wonder what the map would look like in different contexts. In all of San Francisco, what is the threshold for common groups? What about in New York? St. Louis? What would a map of common networks criss-crossing the United States look like? What about one spanning the globe? Do these people with diversely common networks share characteristics that become more salient as we zoom out? Or does the commonality depend completely on the ego’s perspective? How does race, age, gender, nationality, education, or profession affect the makeup of diversely common connections?

Looking at my husband’s network, I find a lot more diversely common connections.

While he has many more connections than me, his networks are much more integrated than mine.

What does this say about his network? Are they generally more social than my connections? Do his connections have a lot more in common than my own? What would an extended map showing the connections of his and my connections look like? Would mine expand broadly like a puddle and his condense into a thick cloud?

It is wonderful to see the variety of connection types captured in our networks. Knowing these people personally, I can begin to guess the logic behind their position in my network. I am eager to understand the science behind these networks and put order to the chaos.

Hackers and entrepreneurs catalyze through the marriage of their networks. Hackers, or creators, are able to reach a broad market that values their intellectual property through the entrepreneurs, or organizers, that recognize value in their creation.

(I’ll define “hacker” and “entrepreneur” in the next posts, so for now let your mind wander to your own definition.)

I picture two contexts where this catalysis is particularly powerful.

Imagine a man in Uganda comes across a library book that possesses a picture of a windmill. The man creates the windmill from the image and establishes it as a power source for his shack. How many people, that cannot walk by and admire his creation, would be interested in owning the same thing? Now conjure an image of another young man from Uganda who has just walked the graduation gauntlet of MIT’s Sloan School of Management. This young man recognizes opportunity in his homeland, and now has the expertise to give life to that opportunity. Who will introduce the windmill maker and the newly minted business maker?

Imagine another young man in Mountain View, California. He works a decent job, sloughing bugs for a software firm in the Valley. But he really gets excited in the evenings, when he works on his latest project: a tool that allows anyone to share vast amounts of data with other users regardless of Internet connectivity. His only frustration is the lack of access to a powerful enough platform that would support his program. An angel investor sits at the dinner table two miles away, dreaming of fueling the next big file-sharing breakthrough. She knows the opportunity exists, but does not know how to make it happen. How will the bug-bouncer meet the angel?

I believe there are key nodes that link the social networks of the hacker and the entrepreneur that activate innovation, pulling the idea from a concept or pilot stage to a full-on market introduction. What are these nodes exactly? Are they highly connected people, people with a targeted expertise, or someone that simply brings the hacker and entrepreneur close enough to meet? Is it one person or many?

The risk from failure – where failure is the unmet linkage of the hacker and entrepreneur – is the stunted economic growth born from missed innovations.

Once we recognize that a connection of the hacker and entrepreneur’s social networks results in innovation (in the sense of an invention meeting demand), we understand that the lack of this connection results in a gap. Is this a gaping hole, or a merely unproductive node that, while linking the two networks, does not actually bring the networks closer? What are the characteristics of the gaps?

These are the questions that I am now exploring.

In addition to deciding what information should be presented, how, and where, AR requires avoiding displaying information in a way that occludes the background detail. View management determines where things are placed in the AR view so as to avoid covering it. In order to determine what 3D objects are in the view and where, a visible-surface determination algorithm is used. As usual, processing power is a key problem with computer vision.

Roles to incorporate in the AR application authoring process: domain expert (e.g. mechanic for a mechanical diagnostic app), story author (translates domain expertise into presentation), content creator (designs presentation), component implementer (animates content), presentation integrator (integrates story, design, and animations), and end user.

Authors begin to discuss their creation of the APRIL language, based on Studierstube. I plan to experiment with Studierstube. To do this, I will need to understand the basics of 3D graphics, openGL, and Open Inventor

To understand openGL, I used resources found at http://maniacdev.com/2009/10/18-opengl-es-resources-that-you-must-know/.

Am currently reading http://www.dpfiles.com/dpfileswiki/index.php?title=Black_Art_of_3D_Game_Programming%2C_Chapter_10:_3D_Fundamentals to understand the fundamentals of 3D mathematics.

Notes from Hall’s “Hidden Dimension”

The structure and usage of the eye can inform design: lines such as trees in the peripheral vision intimate movement, so a tree-lined road causes drivers to slow down whereas a smooth tunnel causes drivers to speed up. Peripheral movement induces a sense of crowding, so restaurants may want to increase peripheral activity for a sense of crowding or decrease it for more personal space.

Thesis Planning 2/27/10

The goal of my thesis is to give AR app architects a guide they can use to build AR apps based on a need, rather than the usual scenario of looking for a use case that meets the technology’s constraints. I am struggling to avoid distraction and deep diving into the details of AR app development. This is the reason I took a step back and reviewed my thesis goals and plan. I now realize the priority of interviewing AR app companies. I have identified Layar, Metaio, Presselite, Mobilizy, Total Immersion, Acrossair, and Eyeply as companies of interest, as well as U Washington, U Santa Barbara, Georgia Tech, and HITLabNZ as research labs of interest. I plan to conduct an analysis of the companies, using techniques such as SWOT, to understand who is doing what and how successful they are so far.

I plan to continue reading about the technology behind AR, practicing AR development, and reading about the human needs behind AR. At some point, I want to incorporate emerging technology investors and marketers in my research. I am also looking into building better business models, design thinking, and opportunity recognition literature at the risk of unnecessarily deep diving into the details.

Was able to access financial data on some AR companies, particularly Total Immersion, via Orbis. Will need to learn how to analyze this data now.

3/7/10

From Emerging Tech of AR book, authors compile XML-based APRIL presentation and interaction language.  They divide the aspects of an AR application into the setup, cast, story, behaviors, and interactions, where behaviors are the actions of the cast and interactions are the actions of the user. They use UML, such as Poseidon, to create a UML state diagram storyboard of a simple APRIL application. See p. 153 for a schematic view of the APRIL transformation process.

“AR as technology” versus “AR as medium” (p. 161)

Need to test out Studierstube and APRIL on a Windows PC (ISchool Lab)

Other sites to check out: http://www.hitl.washington.edu/home/

Will need to download the trial of Adobe Director to test out DART (http://www.cc.gatech.edu/dart/)

Finance and Accounting for Nonfinancial Managers, ch. 6: Need to calculate an “industry average” ratio for AR, and compare this with that of related industries (Internet, mobile advertising, software, social networking sites, etc.). Ratios to calculate (see chapter 6) include profitability, liquidity, operating efficiency, and capital structure/leverage.

The stages [Nils] sees for the evolution of augmented reality:

AR1 Augmented Objects
— Marker and picture recognition
— augmented print and augmented packaging
— augmented POS terminals
— augmented shopping

AR2 Augmented Layers
— Mobile AR, no marker
— location based and compass
— AMOLED (Active Matrix Organic Light Emitting Diode)
— Pico Projectors (MIT)
— cloud computing
— wearable displays
— augmented goggles
— outernet,
— gesture control
— object and face recognition

AR3 Augmented Vision
— active contact lenses (total immersion)
— MARE (Massively Mobile Multi-User Augmented Reality Entertainment)
— MMTRG (Massive Multiplayer Trans Reality Games)
— augmented cinematic games
— pervasive wireless broadband
— semantic search
— intelligent pattern and image real-time processing
— brainwave control and voiceless communication
— mood detection
— real-world avatars

AR4 Augmented World and Augmented Brains
— artificial intelligence
— eye chips
— brain-computer interfaces
— “Web of thoughts”
— five-sense immersion

AR5 Full AR perspective
— neuro enhancement
— full body prosthesis
— “Neurobots”
— telepathy
— artifical brain/brain transplantation
— brain upload and caching
— reverse-engineered brain

I explored the ARToolKit this past week. It was a lot of fun, but frustrating from a mobile perspective. Android and iPhone ARKits are available. I am currently exploring these programs, along with OpenCV (open source computer vision), although they are still very rough around the edges. I decided to drop the FLARToolkit from my to-do list since Flash isn’t yet an option on most mobile platforms.

I’ve begun reading Haller, Billinghurst, and Thomas’s “Emerging Technologies of Augmented Reality: Interfaces and Design”. It’s an interesting book in that the authors introduce it as a guide for interface design even though it focuses heavily on backend architecture. One important quote so far: “the ultimate test of obtrusiveness of a [Head-Mounted Display is] ‘whether or not a wearer is able to gamble in a Las Vegas casino without challenge.’” (actually a quote of a quote from Bass, Mann, Siewiorek, and Thompson, 1997). Through chapter 5, the book seems to provide more information on non-mobile AR. The rest of this week’s reading from the Haller et al piece will focus on AR development environments. In addition, I will be reviewing the ISMAR proceedings from the 2009 conference and UC Santa Barbara’s active AR research (Four Eyes Lab).

I found a terrific resource on the history of mobile AR that I plan to reference: https://www.icg.tugraz.at/~daniel/HistoryOfMobileAR/.

AR UX

I continued reading Hall’s “The Hidden Dimension” this week. Hall is beginning to go into more detail on “perception of space”, namely vision, hearing, smell, touch, and heat. Hall divides these perceptions into distance and immediate receptors. Based on his description in previous chapters of the physiological effects of crowding, Hall connects the perception of space with the sensation of crowding (which can be individual as well as cultural). A particularly interesting point he makes is that active touch (“tactile scanning”) is distinguishable from passive touch (“being touched”). Hall notes that active touch leaves a stronger mental imprint than passive touch, and that more interfaces should incorporate active touch to encourage a more vivid experience. While Hall uses the example of the automobile as a flight from active touch, the opposing experience of riding a motorcycle could illustrate the more intimate experience involved in active touch. The overarching theme of the tactile experience is the level of intimacy involved with touch. Hall states that, “Touch is the most personally experienced of all sensations.” While tactile interfaces present a striking design opportunity, they are also keenly sensitive to the relationship with the user.

Hall closes his overview of perception of space by asserting that, “Man’s sense of space is closely related to his sense of self…” As we explore the nuances of AR experience design, it is vital to understand that space is intimate and personal to the individual. No design standard is applicable here.

What is this all about?

To culminate my graduate education at the UC Berkeley School of Information I will spend the next four months applying my lessons learned to the practical problem of designing mobile augmented reality (AR) applications that are driven by the needs of the user as well as those of the business and the technical stakeholders. My final product will be a public white paper addressed to AR entrepreneurs and architects that will guide them in building an AR application that has a distinguishable market. I will do that specifically by surveying the evolution of AR technology – how did we get to where we are, what can we do now, and what is in development.  Then I will discuss core user needs that must be considered when building any AR app. Finally I will explain how basing your AR application on a foundation of user needs will produce a higher return on your investment.

I plan to post reports of my research regularly on my blog for public consumption. The point of my research is to provide information that is helpful to you, the AR application entrepreneur and architect. I hope you’ll help me by guiding this research in a direction that’s useful for you.

AR Technology

Over the past two months, I have begun learning the fundamental technical issues that face AR developers. AR can be organized into two primary categories that reflect the manner in which the information is recognized by the computer. The first is image-based, with the computer relying on a recognizable pattern via a marker (see QR codes) or the edges of the object (see SIFT/SURF feature descriptors). The second category is location-based, which currently means relying on GPS.

Pattern-based AR is the most widely used form of image recognition because it does not rely on the phone having a compass, accelerometer, or GPS capability. Any phone with a camera can take a photo of a QR code, for example, and be able to access an application based on that image. However, when trying to capture more complex patterns, such as text, environmental factors like shadow or movement can impact the accuracy of the image recognition. Also, image recognition programs rely on a database of images that might match the target image. If the program does not contain the image, it will not be able to recognize the object. Thus while some applications are able to capture, with relatively good accuracy, the patterns of catalogued objects such as books or record labels, capturing dynamic or uncommon objects is much more difficult.

Location-based AR has its own challenges. Location-based AR relies on GPS and, to create a 3D interactive environment, a compass and accelerometer. If the experience is meant to be live, a video camera is also required. GPS is not reliable within about 150 meters.

Other technical issues to explore include internationalization of AR applications (local constraints and tools), platform segmentation, language interoperability, limited processing power, screen resolution, and vision tracking.

My next step in exploring the technical issues of AR will be to program in AR environments, such as ARToolKit, FlarToolkit, and OpenGL. I will also be reading Haller, Billinghurst, and Thomas’s “Emerging Technologies of Augmented Reality: Interfaces and Design”.

AR UX

The AR user experience presents some exciting challenges in itself. Augmented reality is essentially an intimate interaction with your environment. This interaction can be as active as a dialogue with the environment, or as passive as an awareness of the environment. The designer’s challenge lies in deciding which sensors will best perform the desired interaction. Should the program use haptic, visual, olfactory, or other communication methods? One proposal is that we translate lessons learned from the gaming industry’s success in creating emotional and goal-driven experiences into the similarly immersive experience of AR.

Filtering noise is a key issue in any mobile user experience. The user has limited screen real estate compounded by limited time, in most mobile contexts, to explore mobile content. This situation lends itself to layering content, that is to say prioritizing the content according to the user’s context and needs.

User control is paramount in an immersive experience. One way to provide the user with full functionality while performing a service under the user’s control is to layer automation and privacy.

AR developers will do well to learn from the commercial experience of virtual reality. Which VR applications were driven by user need? How successful were they in comparison to those driven by the technology?

Unfortunately there remains a chasm between the academics who are advancing the field of AR and the entrepreneurs who are applying it. An ongoing conversation needs to be fostered here, ensuring that AR applications follow an innovative trajectory that benefits the users and other stakeholders.

A large percentage of AR applications are in the marketing space. AR, when combined with location-based services and social networking, has a chance to revolutionize marketing. Imagine a relationship between a provider and a customer, built on mutual respect and confidence in each other. AR is one step closer to building that positive relationship, rather than a relationship of the passive consumer inundated and force-fed irrelevant and unwanted advertising. If AR user experience designers, developers, and marketers are able to collaborate on building a better customer experience, this vision can be achieved.

Unfortunately AR is not currently available on every mobile platform. It is limited to devices that have some sort of camera at the very least, and other hardware if you want a more interactive and real-time experience. For those who do not have access to this hardware, information provided via AR is not available to them. This could render an unfair advantage in areas like environmental or urban monitoring, where a person is not aware of the pollutants in their neighborhood or is targeted for discriminatory practices and unaware that they are being misrepresented. Universal access should always be a goal of AR developers and designers. Besides, making your application accessible to many platforms means that your market is larger!

My next step in analyzing the AR user experience is to learn more about the culture of space as described by Hall in his “The Hidden Dimension”.

Please Comment!

As usual, please comment if you disagree or have further information on one of these topics. This is part of a process, so please forgive half-baked information by pointing me in the right direction. Thanks!