information visualization and social visualization

. Early work on information visualization emphasizes the power of graphical representation in enhancing our cognitive capabilities to process data and information. By representing abstract and numerical data in graphical forms, it makes the data and relevant patterns more perceptually intuitive and easy to understand. its power has been widely employed for data intensive fields such as scientific and business data processing and analysis . As Card points out, "Information visualization is commonly defined as ''the use of computer-supported, interactive, visual representations of abstract data to amplify cognition"

More lately, graphical visualization techniques have been applied not just on scientific or business data for the purpose of intellectual inquiries and analysis, but also on social data (e.g. social presence and activities) in support of social interactions, especially in the field of computer-mediated social interactions. The latter is generally explored under the term of "social visualization" by Donath or "social proxies" by Erickson and Kellogg. for revealing patterns, structures and impressions, graphical representation is considered to be superior than text-only interfaces.

Issues with social visualization is different from general information or data visualization in a number of ways. First, social data is usually subjective and inexact, and poor choice of graphical representations can be easily misleading and spurious. For instance, Donath et al. () discuss how the color codings tend to suggest meaningful interpretations (e.g. orange conveys a cheerful mood while a dull brown is depressing) and careless coding might introduce confusions. Fisher have discussed this issue in their work. Second, social data is a more sensitive material than general dataset, since it is to disclose information about people after all, which requires designers to be more responsible and sensitive to the potential ethical, political and moral issues that social visualization might introduce in the setting. In particular, privacy factors must become an integral part of design considerations. Third, social data is more dynamic, especially when visualizing real time social data for on-going social actions, which poses great challenges for designers to accommodate unexpected characteristics of incoming data.

the central question, Donath et al. tries to answer with social visualization is "what does an on-line conversation looks like?" They have explored social visualizations, in an attempt to reveal individual identities and convey the ambiance of the site as a way to provide a richer environment for online social itneraction as well as provide insights of the community itself. the They are concerned with conveying the right information that is already there and try best to avoid being misleading and spurious.





. Nowell (Graphical encoding for information visualization) suggests that no more than five to six colors should be used in coding a display and that color-coding should only be used for the information that is most directly relevant.

2. social visualization in actions

Technology is a double-edged sword

What isn't? Stone, Knife, Wealth, Beauty, etc. With no exception, they are all double-edge swords.
What new insights Steven Levey's "Tech's double-edged sword" can tell us?

"The more powerful our tools are, the more dangerous they are when turned against us"- very true!

Lawrence Lessig discusses the phenomenon of internet, how it was like a shooting star, flaring across the night sky and disappearing just as unexpectedly. It suggests that the new laws and regulations is dismantling the very architecture that allows internet a framework for global innovation.

Lessig discusses the notion of commons, the resources that everybody has equal access, and is not controlled and regulated, unlike other private and owned properties. What is the role of commons in our society? What does commons do to a culture? (it reminds me of a particular way of thinking of places: some are designed to distinguish, others are designed to remove distinctions. High profile places targeted one kind of consumers, the wealthy, powerful and famous. Other places, such as a coffee houses are more widely accessible. It is not which is better. I believe we need both to exist. ) for traditional physical resources, limiting commons makes sense, because individuals might over-consume resources and lead to depletion, but there are particular kinds of resources that are not subject to this rule, such as knowledge, music, poems, and now the software. The resources will not decrease when someone gets a copy. This insight of the distinction between software property and others is important!

He talks about the internet as composed of several layers, the physical, code/internet/logical, and the content. Neither the physcial layer (computers, cables) nor the content are resources of commons, only the middle layer is!

"This balance of control and freedom produced an unprecedented explosion in innovation. The power, and hence the right, to innovate was essentially decentralized. "

"This history should be a lesson. Every significant innovation on the Internet has emerged outside of
traditional providers."Patent regulation, although was designed to spur innovation, maybe in the end do harms to innovation. Because it is taxing to apply for patent, which means only those who are powerful like big companies can afford the tax to do it. "The law becomes a tool to assure that new innovations don't displace old ones"

"the right to compensation shouldn't translate into the power to control innovation."

Design and use issues in HCI

HCI, as a field centrally concerned with user experiences with computing systems , has long been taking the intersection between technology and human being as its central focal of inquiry. However, the relationship between technology and human being has been reduced to relationship between technology and users, the relationship between designers and technology, and relationship between designers and users are largely under explored.

Daniel Fallman discusses three accounts of design: conservativeness, romanticism, and pragmatism. While conservative account as derived from science and engineering tradition, emphasizes a transparent , rational and rigid process which lead to design solutions to preset problems; romanticism, following artistic practices from poetry, sculpture, music, etc, emphasizes individual imagination and creativity, which put more attention to the final product and individual designer, and largely regard the design process as a "black box"; pragmatism, rather than science or art, takes the form of a hermeneutic process of interpretation and creation of meaning. It views design as a dialogue between designers and the settings crammed with people, cultures, values, materials, etc., where designers engage with the world, make do with what is available in a specific situation, define the materials and tools at hand by its potential use. (I feel it was an extension of phenomenology from issues of use to issues of design). It operates from the available means, and treats them abstractly, by seeking to determine and redefine the roles they can play in a given situation by entering into a dialogues with them. So rather than leaning on theories and methodology for guidance, pragmantic accounts acknowledge the pre-reflective knowledge of everyday life and work as the main elements of knowledge. Designers in this veiw is thought of as a "self-organizing system" with constructutive as well as reflective skills.

He argues that none the three aspects alone can fully account for what design is. We can not treat it as a fully scientific conduct, nor black-box, nor an unimportant practical bustle with reality. It should not be treated as a disciplinary mix, existing in between science and arts. Rather, he argues, design is a tradition that guides action and thought. "it is a tradition f pro-activity; and active stance. It includes the archetypal activity of sketching by which designers becomes involved in the reflective conversation needed to bring new artifacts into being." "the role of design in HCI is thus to be found in the act of trying to unfold a coherent whole, from the various bits and pieces gathered together in the process of research." "Field work, theory, and evaluation data provide systematically acquired input to this process, but do not by themselves provide the neccesary whole. For the latter, there is only design."

However, I am not convinced how distinction between design-oriented research and research-oriented design can be useful.

Shall we consider deployment as design, shall we consider use as design? design is a very fuzzy word.

Phoebe Sengers and her colleagues propose for reflective design. They argue that reflections on unconscious values embedded in technologies and every practices it supports should be a core principle of technology design. In their definition, reflection is referred to as critical reflection, which bring unconscious aspects of experience to conscious awareness, thereby making them available for conscious choice. Therefore, reflection, or critical reflection, is important for "individual freedom and our quality of life in a society as a whole", because it allows us to consciously think about attitudes, practices, values and identities that we might unconsciously espouse.Furthermore, since our everyday experiences with the world are shaped by our unconsciously held assumptions as part of our identity and the ways we experience the world, critical reflection can open opportunities for us to experience the world and ourselves in a fundamentally different way.

Senger's reflective design encompasses a much broader agenda. It include reflections for designers to rethink about their design practices, what values are embraced in their design choices, the role of users in the design process, and also provoke reflection on users what role the technology play, what is the relationship between technology and every lives. Much of their work of reflection design focusing on challenging the role played by technologies in our lives. Thus, they propose to embrace reflection not just in the final products, but in the whole design and evaluation process.

Furthermore, sengers also argues for an integration between reflection and actions, "critical relfection is effective only when it is immediately folded back into our experiences, actions, identities, and practices, rather than intellectual practice separate from actions." Thus social information, information about user presence and activities and patterns over time, should not just used for post-hoc reflection and evaluation, as traditionally favored, but should provide an "ongoing opportunity for everyday, open-ended reflection for the users during their own activities".


Phoebe Senger is very inspiring. It inspires me to think, traditional conception of location-based computing centered around locating, tracking, coordination, and etc. It is concerned with the accuracy, granularity, and ways of locating people. However, location could mean much more. It does not just indicate activities, but also collective and historical whereabouts can convey an atmosphere of the collective. Are they all together, are they isoloated in a quiet space? how often do they use the public space? will they tend to spend a lot time at certain play, or move around? How busy the faculity life is, they might be able to people's sense? people's collective use of the space overtime is a narrative of people, and vsualization provide a means to engage people to reflect and think of these questions.

Also by providing an easy means for people to opt in and opt out, it provokes them reflect on presentation, not just merely disclosure. to think of what does the place mean for them. when exposure is too much, how do they adjust to level of exposure. By providing simple means for them to supply place name and activity information, give them an chance to play with it, give them opportunities to shape the use of the technologies.

reflective design is a framework and a set of design strategies derived from a set of critical approaches: participatory design, value sensitive design, critical design, ludic design, critical technical practice, reflection-in-action. They argue, design should incorporate reflection as its fundamental principle.


reflect design argues for incorporating dialogue between designers and users in the

Bill Gaver suggests using ambiguous output to discount the interpretations made by sensing and learning in ubiquitous computing systems. Due to the inherent heterogenous and idiosyncratic nature of human activities, computer are relatively incapable to understand the full meaning and activiites and context. Therefore, they explor using embiguous outputs as a tactic to undermine system authority in favor of user's interpretations, stimulating reflection in ways.

Design for engagement

Traditional system design and HCI favor accuracy and efficiency. These are values derived from task oriented and engineering tradition of computing design that are largely originated and take place in work places. As computers continually spread into other places of everyday lives, where computers are embedded in settings not just physically different from work places but also culturally, socially and historically different. It challenges these underlying values and approaches inherited from engineering tradition.

Built on Philip Agre's notion of critical technical practice which proposes that practices of technology design should incorporate philosphical and critical reflection , Pheobe sengers and her students propose a program called "cultural embedded computing", with its aim of building technologies not just for people to use, but also for people to think about technology and themselves. In this program, they turned the role of user studies from straight technology evaluation towards sociocultural research, where the goal was not just to improve understandings of people and technology, but also encourage participants to rethink their own experiences in light of their participation; they also designed systems to trigger people to interpret and reflect on the relationship between machine and emotions (the influencing machine), and interpret and reflect on a group;s ongoing emotional experiences.

Therefore, the way of measuring the success of system turned away from measuring efficiency towards concerns with how people worked with the display and what meaning they attributed to it. So the technology is not just a tool to inform affection or others, but what experiences that display can evoke.

Adoption and use issues of new technologies

Bradner, Kellogg and Erickson's study of the adoption of the "Babble" system illustrates again how local cultures together with technological characteristics can impact the adoption and use of technologies. They also proposes a number of notions and frameworks to understand the adoption phenomenon, such as critical mass (they pointed out that not just the absolute number of participants are important, but also different types of participants for different types of communicative practices - while announcement practice requires a large audience, "good morning" greeting might only need two people), social affordance (whether the interaction between local culture and techological characteristics and afford certain types of practices), and ecology. They illustrate that in the shared culture, people can easily infer about what is going on with the small social cues in the babble systems.

Design for privacy in ubiquitous computing environments

Bellotti and Bly discuss various ways technology breaks down social communication conventions and mechanism that are taken for granted in everyday world, and suggests that technologies result in disembodiment of context from and into which one projects his information and and disassociation of one's actions. For example, when computer mediated, a lot of social cues are attenuated or we might not realize that our information is visible to others. For disassociation, it might be that, for a shared workspace, only the results of our actions are shared, not our the process of actions itself. Considering these issues, they propose a framework for designing and evaluating privacy issues in ubiquitous computing systems.

It seems while RAVE system is well accepted in this specific culture, it was highly suspected it might very likely be subject to sinister use in other cultures.

It is interesting that they discuss the privacy mechanism itself might introduce new privacy issues. for example, in order to inform the visitors in the commons area with cameras installed who are connected and are able to see them, one solution is to project images of those who are connected, which then makes individuals who connects to the common area feels intrusive. it is an interesting dilemma, when you introduce new solutions to other problems, these solutions introduce new problems to solve.

They mentioned issues of awareness and privacy: while providing too much awareness information of others people's activities and availabilities might seem to be intrusive, so is too little awareness information - it can cause intrusion too when one can not tell easily the availability of others.

Culture and control in a media space

Dourish discusses the technical-social relationship in regulating access and accessibility in media spaces. He suggests there is a social-technical continuum along which mechanisms of regulation is located and enforced. While some media spaces such as Xerox PARC's don't have explicit technical constraints on media space access and use, it has a explicit "sign up" step in order to be part of the media space. This way, it is subject to the soical practices and norms to govern acceptable media space use. Others such as the Godard in Rave, which provides a set of mechanisms for users to set access and accessibility among a set of services including a sound based feedback mechanism. In this way, the control is largely technical to reduce the misuse of media space technologies. At the same time, he emphasizes even in Godard which has explicit technical constraints for access and accessibility, it is still subject to social regulations, for example,social pressure may mitigate against refusing video access to superiors, or the feedback on the other end will render constant "glancing" into others' offices inappropriate.

Surveillance vs. Souveillance

Steve Mann distinguishes between SURveillance which means "eye-in-the-sky" and SOUveillance which emans bringing cameras from the heavens, "down to earth". ("Sousveillance")

Surveillance refers to situations where people of higher authority such as government, department store owners, safety guard and the like watch and monitor the citizens, shoppers, and suspects from the above. The authority is often seen as "Godlike" rather than down as the same level as individual parties. He refers the capture of multimedia (audio and video, etc) content of individual activities by a high entity as surveillance.

As an inverse of surveillance, he suggests the notion of souveillance, which refers not people of high authority looking down, but individual parties looking up and looking around, such as citizens photographing police, shoppers photographing shopkeepers (looking up), or individuals recording scenes around for personal experiences (looking around), etc. So it is both a reversal of hierarchical structure, and human-centered experiences.

Surveillance//souveillance
God's eye from above//Human's eye view, down to the earth

Cameras' usually mounted on high poles, up on ceilings//camera down to the earth, at the ground level

eyes in the sky//eye in the eye

watch from above//to watch from below

architecture centered, cameras mounted on or in structures// human centered, camera carried on or worn by people

recording made by authorities, remote security staff // recording of activities made by the participants' of activities

In most of states, it is illegal to record other parties' phone conversations//in most of states, it is legal to record conversations when you are a party

recording is usually kept in secret//recordings are made public (e.g.on the www)

processing is usually shrouded in secrecy // processing is based on open source

originated from panoptic in the context of prison, surveillance tends to isolate individuals from one another, while setting forth a one way visibility to authority figures// it has the community based origin, it brings people together, and makes the large city functions more like a small town, with the pitfalls of gossips but also the benefit of sense of community

privacy violation tends to be unnoticed, unchecked and not be self-correcting//privacy violation is usually immediately evident, and tends to self-correcting
it is hard to talk to people behind the surveillance cameras//there is chance to talk to people behind the souveillance cameras
when combined with computers, surveillance tends to rely on cooperation with the infrastructure around us//wearable computing, doesn't require cooperation with the environment
the locus of control tends to be on the authorities// it is possible for the locus of control to be more distributed, to rest with individuals

The Transparent Society

David Brin's book "the transparent society" argues for a more transparent and open society, where we will not just be watched by authorities, but we will also watch the authorities and each other. The key is a two way watching, or reciprocity or mutuality. His basic argument is that, the transparent society will encourage criticism, which is the mechanism for civilization and progress, and what we need is a courage to face criticism. So while some argue for shields and masks for privacy, he argues for transparency and openness for privacy.

The problem is whether "mutuality" can be actually achieved in reality?

Symbolic interactionism

According to Herbert Blummer, symbolic interactionism is a distinctive approach to group behavior and human conduct from other more traditional approaches in sociology and psychology. It rests on three premises:

1. human beings act towards things on the basis of the meanings that the things have for them. So our actions are not merely driven by some psychological factors such as attitudes, conscious or unconscious motives, or incentives or external social factors such social roles, statuses, social demands, social norms, cultural prescriptions, etc.

2. means arise out of social interactions that one has with one's fellows. So it is not that meaning is inherent in the things and just emanate from them as held by the realism philosophy; it is not meaning just as an expression of human subjects' psychological elements such as sensations, feelings, ideals, memories, motives, etc.

3. the meanings are handled in, and modified through an interpretive process used by the person in dealing with the things he encounters, so the use of the meaning is not simply arousing and applications of already established meanings..

The wisdom of crowds

How can a group of stupid individuals that only think locally (within their own specialization and local knowledge) result in a solution that can serve the overall common good for the collective as a whole? "the wisdom of crowds" by James Surowiecki convinced me with numerous examples that that is possible. But why are there many examples and arguments that a group of people is more stupid than individual? James suggested that three conditions are important to ensure collective wisdom: diversity, independence, and discentralization. I think that is very insightful.

Diversity is key to collective wisdom. First, because it provide alternatives for options. Like many industries in history, at the start, there are always many kinds of design, many kinds of technologies, and many kinds of styles produced many companies trying to create a market. However, in the end, usually one a very small of them remain. None was able to pick the winners beforehand, but "what makes a system successful is its ability to recognize losers and kill them quickly." (p.29). Second, the diversity of options are not enough, but the group has also to be diverse because only so we can have different perspectives, and diversity has its value on its own right. "Homogeneous groups are great at doing what they do well, but they become progressively less able to investigate alternatives" (P.31)

"Ultimately, diversity contributes not just by adding different perspectives to the group but also by making it easier for individuals to what they really think...diversity helps preserve that indpendence, it's hard to have a collectively wise group without it" (p.39)

Another key condition is independence. It is interesting that he discuss independence that we as autonomous beings with we as social beings. Although he recognizes the social nature of existence, he argues that "the more influence a group's member exert on each other, and the more personal contact they have with each other, the less likely it is that the group's decisions will be wise". that is a very strong point, and kind of work against what we believe in social computing, which is based on the assumption that, we are fundamentally social creatures, and if we make social information visible, if we encourage social interactions, we can have positive results either the enhanced connections or informed decisions. He mentioned the information cascade effect, where people who do not have complete information will follow and learn from others, however, bad things happen when at certain point, people become stop paying attention to their own knowledge and just blindly imitate others, this is when information cascade stops being informative.

The final condition is decentralization. Decentralization foster and is fed by specialization. it also means only people know the local and specific situation and come up with the best solutions. The key is a balance between making individual knowledge globally and collectively useful will still allowing it to remain resolutely specific and local. "A decentralized system can only produced genuinely intelligent results if there's means of aggregating the information of everyone in the sytsem"