CHI 2003: How to Make Group Environments Social

"Designing Visualizations of Social Activity: Six Claims", by Tom Erickson of the IBM T. J. Watson Research Center, does what it says on the can. Drawing on extensive experience in designing visual representations of groups in online environments, Erickson offers the claims as a starting point for design work, and to drive critical discussions amongst design stakeholders.

'Personal computers are fundamentally social tools,' he says. 'From their earliest uses for the production of spreadsheets and other documents, through the rise of email as a ubiquitous form of communication, computers have been used to mediate communication among people. The current escalation of instant messaging (IM) technologies as channels for social and business talk is yet another illustration of the social nature of computing.'

However, IM and the like bring a new level of social 'presence' into computer use, supporting both unexpected encounters and interruptions. This brings added import to the question: How should people be represented in computational contexts?

A social visualisation, he explains, is a visual (or sonic or other perceptual) representation of information from which the presence, activities, and other characteristics of members of a social collectivity may be inferred, and, by extension, can provide the basis for making inferences about the activities and characteristics of the group as a whole. Examples of social visualisations range from minimalist work to the more mimetic representations found in 3D virtual environments.

. Everyone sees the same thing; no customization
'Design stakeholders often suggest that users should be able to customize social visualizations. For example, a user might wish to make themselves invisible to other users. While this is, at one level, a reasonable request, it is contrary to our aims. An important aspect of the power of a social visualization is the knowledge that everyone sees the same thing. If I see something, I know that you see it as well and that you know that I know. It is this mutuality that supports people being held accountable for their actions, and that leads to useful social phenomena such as feelings of obligation and peer pressure.'

. Portray actions, not interpretation
'Social visualizations are often designed with a particular usage situation in mind, and thus it seems natural to surface the intended meaning of an activity in the visualization. However, systems often end up being used in unexpected ways, and what was supposed to be a feature for increasing ease of use (for the intended situation) becomes a bewildering or, at best, irrelevant feature. Instead, we recommend minimizing the amount of interpretation that is built into the system; let the users interpret—they understand the context better than the system ever will. Our users have proved much more adept at providing appropriate interpretations than we could ever have built into the system. Even if it were possible to somehow accurately build interpretations into the system, we suggest this is a bad idea because:'

. Social visualizations should allow deception
'In the course of our face to face interactions, it is often the case that we go to considerable effort to project impressions that don't represent our underlying feelings. We may feign interest, nod understandingly when we are baffled, and act pleased to meet people we loathe. These are vital social skills, and the last thing a social visualization should do is undermine them.'

. Support micro/macro readings
'Whenever possible, a social visualization should be built up out of many small components which persist. Ideally, over time, information will accrete into recognizable patterns at multiple levels.'

. Ambiguity is useful: suggest rather than inform
'When we discuss social visualizations with engineers, a common concern is how well they scale: this works well for a dozen people, they say, but how about thousands? Our response is that accurately presenting information is not the point of a social visualization; its primary role is to provide grist for inferences, and, in fact, it is less important that the inferences are correct. Our users have proved very comfortable with making best guesses from incomplete information. Thus, it is OK to distort activity, to magnify small amounts of activity, and to dampen large amounts of activity; for example, it is much more important for users to be able to tell whether there are 3 or 7 people present, than whether there are 103 and or 107 present. Ideally, the ambiguity of the visualization should be clear to users.'

. Use a third-person point of view
'Although it might be argued that user's do not need feedback on their own activity since they know what they're doing, our experience is that this is quite important. People learn what elements of the social visualization mean by watching it over time, and, particularly, by seeing their own behavior reflected in it.'

These claims are derived from work with "Babble", a persistent chat system that supports everyday interaction among members of medium sized groups. Erickson refers to them as claims, rather than principles or guidelines, because, while though of their validity, he is less confident that they apply to any collaborative system. And, he adds, in discussions with design stakeholders, framing them as claims — rather than guidelines (which are to be followed), or principles (which are "true") — can lead to interesting and useful discussions of the roles which social representations can play in a system design.

Online videos: Engaging your users
Source: Webcredible, 6 September 2008
 
A guide to using online video on your site to offer more enticing content and provide a more compelling user experience.

The Office of the Future?
Source: Occupational Hazards via Ergonomics in the News, 5 September 2008
 
The office of the future is less about technology and more about physiology.

This year's 10 best-designed Application User Interfaces
Source: NNg, 4 September 2008
 
Nielsen Norman Group recently announced the winners in its inaugural competition to find the 10 best-designed application user interfaces of the Application Design Annual 2008.

Caroline's Corner: Buttons on Forms - where to put them, and what to call them
Source: Caroline Jarrett, 3 September 2008
 
Should you put the ‘OK’ button to the left or the right of the ‘Cancel’ button? Like so much in forms, the simple answer isn’t really appropriate. And yet, who needs another ‘it depends’?

Three Usability Gurus
Source: avangate blog, 2 September 2008
 
A personal view, but lots of good links, especially if you're new to usability.

Ofcom publishes report on Usability Event
Source: Ofcom, 1 September 2008
 
Independent regulator and competition authority for the UK communications industries, Ofcom, has published its report on usability issues.

Get Fit for Remote Working
Source: UN, 30 August 2008
 
BT Business has announced the launch of a practical guide 'Get fit for mobile working' designed to help mobile workers avoid back, neck and arm problems.

Site Visit Interviews: from Good to Great
Source: User Focus, 29 August 2008
 
For those of you for whom the Basic Introduction to User Interviews wasn't quite enough.

Six Metrics for Managing UI Design
Source: Russell Wilson, 28 August 2008
 
A proposal of six metrics to be used for managing a user interface design department.

Don't Judge a Form by its Cover
Source: Formulate Information Design, 27 August 2008
 
The saying "don't judge a book by its cover" reminds us that looks are deceptive. It turns out that this idiom applies to forms too.