Justin's visit has come and gone; the kids learned a bunch from him, and we got to talk about work, school and woman. I also got a small bit of inspiration from seeing his Portfolio, and took it upon myself to revamp some of the ugly portfolio pages on smallbear.org .. As I was looking through old CDs, I came across some funny pictures of J from college :) Also found some pictures of old friends from school (some people just don't change that much!).

Some images from Portland are up; CHI was great, but it sure is nice to be home :)

Kevin's photostream also has some interesting shots.

(at CHI05)

Social Behaviors

Values at Play: Design tradeoffs in socially-oriented game design.
Mary Flanagan, Hunter College.

Values investigation. How can we understand what values are with regard to game design? Values - ideal of a world filled with technologies that are excellent according to standard dimensions of functional efficiency. Want to embody values to which societies and their peoples subscribe. Generally we think of values as price point, or efficiency, etc. What are values regarding society, when thought of in light of software and game design? We think of privacy, security.

Design, then, is a critical junction, and have a lot to do with our culture. If I am a designer and I represent a certain set of people one way or another, what I do trickle down throughout other forms of a media. Only beginning to be interesting in game design.

Challanges, then, for embodying values in design include epistemological issues (how do we create something in software, such as technical/philosophical/empircal modes of inquiry); ultimately, its difficult for designers to understand and create a framework to systematically understand and apply our findings from project to project?

Methodological Framework. Constituted by three activity types, dynamically and iteratively related: Discovery, Translation and Verification to understand socially conscious gaming.

Prior work in this field includes Nissenbaum, Howe, and Flanagan; Case study is RAPUNSEL (www.rapunsel.org). Three year project to develop a game environment for teaching programming to middle-school aged girls. Teaching girls how to program in Java. Goal is to create a self-teaching environment. 11-13 year olds!

Simultaneous endeavors: character design, world design, game design, interaction design, animation control via code, multiuser backend, coding, scaffolding, etc.

Constitutive Activities: Discovery. values in functional definition: "to address gender inequities and address a sector overlooked by the software industry". Equity is a larger value that comes out during the discovery process; we can reward for unusual things, would apply to many kinds of learners (originality). This can be thought of as autonomy; non-stereotypic character representations! Learning styles. Finally, biophilia - engaging with life like characters; and subversion.

Biophillia: Design partners preferred human characters. Humans and non-human animated characters are not alike; many of the kids are anti-everything else other than humans (such as animals, robots, etc). Adults tend toward polygon figures; kids want things that are much closer to humans.

Subversion: 60% of online SIMS players are woman and girls. Neopet's. Two games that are inspiring with regard to the target audience; But there's a lot of subversion in this type of activity. People do subversive things that beg all sorts of questions about game design, especially for children. Victorian doll play culture in the 1880s - people would kill their dolls, trying to kill them and bury them in the back yard. Doll manufacturers began selling caskets for dolls! 10 and 11 year old girls play grand theft auto; they almost all have X-Box and Playstation, but few have computers.

Informally, the children "find their favorite car", they "steal it", and then they "drive". Almost half of the kids playing GTA play it in order to drive around. These are kids that live in housing projects; they may not actually have a car eventually; driving becomes an outlet or escape.

Constitutive Activities: Translation. operationalization - gender equality. Implementation. The game mechanics could represent the value through cooperation. Thus,the game mechanics have as many value implications as goals. Code sharing, then, becomes a way to enhance cooperation; they used P2P. 7% of programmers in the US are woman; numbers earning bachelors have decreased. Yet woman account for 43% of all computer gamers in 2004, and woman account for half of all game purchases. Additionally, 60% of SIMS players are woman.

Character representation: What design partners wanted. They wanted combinations of popular culture icons.

Wild.

Do people trust their eyes more htan they ears? Media bias whlie seeking expert advice. Jens Riegelsberger, University College London, UK

Looking at the issues of trust and the lack of trust online that has prompted much research in HCI. Trust in e-commerce was first found; also thinking about trust in online advice. High trust vs. well-placed trust. Systems can enable "well placed trust".

Trust is required in the presence of risk and uncertainty. Trust is also based on assessments of motivation an ability. Lie detection, or cooperation in virtual teams are both factors that deal with ability as well as expertise.

Existing literature allows for two predictions. One that there is media bias - a richer representation increases trust. There may be a truth bias. Cooperation in virtual teams find that richer representations have higher trust.

Second predictation may be discrimination. Richer representations allow for better trust assessments.

Created an experimental approach to oppose the two predictions. Contrasted text only vs. rich media advisor. They also contrasted expert and non-expert advisors. Measure trust under financial risk - observed behavior with regard to financial ideas.

n=160; 4 rich media advisor conditions; 2 expertise conditions. Thus 4x2 with 20 each subjects. 8 between subject conditions.

Measure: Advice seeking. Advice seeking as an expression of trust; questions were pre-tested for difficulty. Performance in the quiz detremined pay (up to $26+5). On each question, only one advisor could be asked. Advice seeking = advice sought from one advisor/advice sought in total.

Predictions for Advice Seeking; Prediction 1 above would imply equal bars; Prediction 2 would imply opposite bars.

Results within subjects: Clear preference for expert advisor in avatar and photo-text representations (evidence for prediction 2); Less clear when comparing video and audio.

Results between subjects: Strong effect of expertise, and weak effect of media representations (video advice sought more than others combined, avatar advice sought less than everything else combined).

Effects of risk: Rich media advise seeking increases with risk.

Conclusions: Expertise can be detected in all presentations; audio and in particular video increased preference, and avatars are trusted less. An increase in risk leads to higher preference for rich media representations.

Homophily in Online Dating: When do you like someone like yourself? Andrew Fiore - MIT.
http://web.media.mit.edu/~atf/

Used for finding casual and serious romantic partners. 40m unique users in the united states used online dating sites. So what are all of these people doing?

Similarlity is often associated with attraction. Values, demographics, attitudes, attractivenss. Effect might me more nuanced - similarity to what you like about yourself.

Anonymized eight-month snapshot. 153,942 completed user profiles. Messaging, who contacted whom, when, how much, and who replied. Dyadic messaging. large data set; 111,000 distinct contacts (where one or more message were sent between two users). Only 21 percent were reciprocated.

Analysis of similarity within dyads?

Similar how? On what dimensions are people communicating likely to be the same? Religion? Hair color? Categorical descriptors: Marital status, wants children, etc.

Conclusion: People tend to communicate with people with similar interests, consistently, regardless of the characteristic selected.

Scott Carter: The role of the author in topical blogs.
Berkeley.
http://madpickle.net/scott/

How do bloggers value a post?

Blogs are a new internet based communication; blogs have gone from nonexistant to 8million in just about 8 years and are doubling every five months. They are also starting to enter mainstream culture; bloggers were able to enter pressrooms.

Weblog, we blog, or blog. A set of frequently updated posts, usually linked to other blogs. Some sort of social component, including commentary and linking different posts or blogs together. There are increasing number of categories within which they fit; personal, topical, mobile, etc.

Topical blogs, specifically those dealing with law and technology, are focused upon in this talk.

Motivation vs. action. What motivates people to do something, and then what did they actually do? Self reporting captured the motivation, and the action was captured with the text of the blogs.

Looked at 8 law and technology topical bloggers over a year. Used a variety of methods to understand their methods (interviews, diary study, text analysis). Who do they think their audience is? Diaries kept the context of posting; what motivated them to post. Took the form of a webpage that they would fill in every time they posted. Text analysis that analyzed the amount of links, blockquotes, word counts, etc.

Study results:

"A good blogger should be like a beat reporter; using insider knowledge to uncover and communicate a story". Reporter was more respected. Most bloggers thought this was the main thrust of their story; something exposed that wasn't present in mainstream media. the role of the columnist was more common. About 24% of all text present was in quotes, grabbed from popular news. There were roughly five links per post, mostly to other blogs. Considered a post was succesful due to popularity.

Bloggers value a post through report and popularity, but reported posts are rare.

Giving the Caller the Finger ... Collaborative Responsibility for Cellphone Interruptions
Stefan Marti and Chris Schmandt, MIT Media Lab

People love mobile phones but hate interruption. Context aware phones are great but brittle. Socially aware communication agents that monitor calls, converse with callers, and determine the owner's social situation. Create a subtle alert for co-located others, and allow others to influence the agent.

Technology is wireless; use animatronic embodiment of an agent,giving a physical presence; there is a conersation finder badge, and a vibrating finger alert.

Conversation Finder Badge senses locla speech/silence, and broadcasts start and stop messages. Low power radio; listens to other badges, and determines conversational groupings.

Finger Ring has a vibrating pager motor and a single button for user input. Vibrates when anyone in the group gets a call. Anyone can veto the call by pressing the button; the agent then sends the call to voicemail.

(at CHI05)

Design Expo
4:30pm pst

The Stakeholder Forest: Creating an Expense Report.
Jonathan Arnowitz. jonathan.arnowitz@sap.com

Enterprise application that is used across multiple industries to manage expense reporting - travel or otherwise. specifically targeted at large companies, those with larger than 15000 employees. automating policy enforcements.

The business case is a call to action: A phonecall from a CEO to the CEO of peoplesoft; the call was that one has purchased a product and the users reject it, because the user experience was poor. The goal: the elusive intuitive user interface. The challange was those of politics: enterprise software, and the high profile nature of the corporate initiative. There would be great scrutiney of the results that were presented. TOE - Total Ownership Experience; the overall usability, integration, etc of the product. Limitations included technical issues as well as resource issues; the toolkit is proprietary and has limitations as to what can and cannot be done. There is also a resource allocation issue; have the ability to do a lot with a little. Limited headcount in order to impliment this product. Product maturity implied that there was an existing architecture underlying the items; the tools are tied to the underlying architecture, creating a difficult challange.

User Experience Process; full range of design and usability activities. Took a holistic approach, involving field research with 28 participants in an onsite and remote task analysis, including 30+ user interviews, as well as customer focus groups.

Developed a variety of user interfaces; specifically, one created for experienced users and one created for short term users. Went on-site to see information in context; also did a fair quantity of remote testing.

Usability testing with 90 users particiapting in over 9 test sessions; a baseline and follow-up testing,

Successes; usability improvements, business improvements, marketplace results - top 25 in revenue generation (out of 160 apps) and top 10 for all FMS (59 apps); 12 months best seller in its group.

Vista - www.usability.pl/vista
Interactive Coffee Corner Display

People working in organizations are interested to know what their colleagues are doing, but have little chance in following up and finding out about what those solutions are. Many people, however, spend a lot of time in the coffee corners or watercoolers; they exchange details about their lives in that space. The idea, then, was to combine these two approaches and try to "sneak" this informaiton about colleagues or projects in a transparant fashion.

Identified 5 design requirements.

Vista should .. have an interface supporting walk up and use; present information about ongoing projects and people working in the same dpeartmnet; reuse existing information and database to the maximum extent; give follow-up possibilities, allowing people to research after the fact.

The best form for this was a large display. The first iteration was a video prototype, to see if people understand the concept. Floating boxes show projects, and allow people to walk by and deside if they want to get more information. Iteration two involves a testing showing size and movement of boxes, as well as the orientation of the display. Wizard of Oz the 3rd test; the 4th test was a real prototype.

Very interesting idea; simply display of ideas and people, allowing others to gain a global (yet random) perspective of their organization

Looking back to plan AHEAD: Exercising User Centered Design in emergency management. Leo Frishberg - Tektronix, Inc.

Carpenter in the facilities department in one of the world's biggest airports, and you are now in charge of the counter-terrorism excercises in the organization. Usability tests - with extreme risks. Over the span of 18 months in 1995, Cliffside software worked with customers to find the issues to assist novice and excercise developers in developing emergency response excercises.

Company tried to learn requirements about customers; was considered risky, as they knew little about the end users. The design team was 3 people; software developer, marketing, and design.

- Design influences include a lack of common standards and vocabulary.
- Most customers had no or very limited computing equipment
- Excercises are context/organizationally specific
- Excercise development is highly collaborative. Even the most trivial element - such as a building evacuation - may require collaboration.
- 3-Ring binders are the preferred artifact.

Why is there a presentation about something that is ten years old in CHI2005?

Maximize the Experience: minimize the extraneous. Keith Saft - Palm One.
The guideing principles for designing at Palm is about the size of a matchbook. How can one reduce the amount of information and elements on the screen to make things effective? Screen resolution is generally around 320px x 320px or 320x x 480px, so minimizing extraneous information is critical.

Six month development cycle generally used; product definition => flow => UI => Implement => Usability => Acceptance. Ship hardware every six months.

Prioritize features based on what users want - looking at the important and frequent tasks.

Organizing the experience. Start with the center press; relating to the buttons, the on-screen graphical issues; then moving towards menus, and finally towards settings. Think about the user experience in a spiral or ring-like manner.

Finding information. Cropped dead space to remove non-essential information in the thumbnail.

Designing the "World as your Palette" - Hiroshi Ishii, Kimiko Ryokai, Stefan marti
Where does ink come from? Ink flows in one direction. We aren't really aware of where the ink originally came from. What if we could suck the ink from anywhere in our environment?

Usually pick up ink from paint cans or ink bottles, such sa plants, rocks, or clay minierals. What if we can take attributes directly from physical objects and paint with them? We can also begin to use movement too.

(at CHI05)

alt.chi

Urban Pacman.
Sharing the Square: Collaborative Visiting in the City
Streets.

Study of a collaborative city visiting system in use; describe how the system was used for shared sociable experiences, intending to build enjoyment.

The motivation for this collaborative tourism system came from observational studies of tourism and city visiting. Tourists are bombarded with information, and ultimately they have too much data. Most computer support for tourists provides more information; perhaps there simply isn't good information.

How can we support collaboration between tourists?

Collaborative museum visiting systems between online and physically present museum visitors. Visitors using the web, VR and actually in the museum can share their visits. One set of participants confused them with the behaviour; they raced across the room, "mucking around" - the fun of using a system. Trying to beat the online users. A fun playful (but wrong) way of using a system.

Redesign with a focus on flexibility. Minimize reliance on prepreapred content and allow for users to generate their own content - thus supporting play of the system in a flexible non-modal configuration (mobile and outdoor use).

Led to the "George square" system. Groups of indoor and outdoor visitors can share a city visit over the internet. A system run on a tablet, outdoors, can merge with users indoors using a PC. They can share voice, location, photographs, and recommendations.

Trial using the system found that visitors form a division of leisure. Indoor users browser the web, where outdoor visitors took photographs. Indoor visitors asked for photographs, where the outdoor visitors asked for information.

Collaborative taking as well as sharing of photographs; visitors took photos together, chatting about the photos as they took them.

The social interactions were a mixture of high and low culture; thus, interactions made use of the system as local resources forinteraction. Talk for the sake of talk.

Does Pacman Need a Helper? Analyzing Experience of Physical and Social Interactivity. Humans, as social creatures, find physical interaction, touch and human to human presence essential for the enjoyment of life. (Bowlby) If we want to create new types of gaming, this is an essential element that we should try to incorporate.

Real world gaming; the growing trend is to require human physical movement as part of the interaction (dance dance revolution). Physical games are stirring up a lot of research interest.

The aims of the reserach are to connect people via new channels of interaction. Human Pacman is a physical computer game with many types of sensing and interaction. Physical interaction with surrounding (collecting bluetooth cookies), physical interaction with players (capture event), physical-virtual interaction (collecting virtual cookies) and social interaction with online players.

Collaboration with a helper gives a broad (forest based) view of the whole of the system rather than the smaller, tree level view. User studies aimed to examine the importance of social and physical interaction.

Short Paper Madness

Threads of Recognition - using touch as input with directionally conductive fabric. Smart fabric in a wearable computing garmit. Sensual - it begs to be touched - the fabric surface is a tactile interface. Playfull, intended to use gestures mapped to the garmets. Interaction is around selecting and directing; self/self, self/other, self/group. The wearer directed biometric data through the hems, aspects, etc of the garmet.

Recognition is based on effort-shape analysis, a movement language for observing and defining language. 1920s. Effort-shape is based on weight, flow, space, time. These four efforts are combined to create effort qualities; they are computationally related to each other and can be mapped. Extract paramaters based on touch, which can be measured through the surface.

The threads use optical fibre arrays, mapped to a pocket PC and cell phone, then embedded in the garmet. Prototyped as a passive conductor, replacing cable. Conductive fabric used to design the user experience, thinking abotu small swatches or foam pads, carbin impregnated, using low-res to explore the aspects of interaction. sewn with hands, beautiful, and beg for the eroticism of touch.

Dishmaker: personal fabrication interface. MIT.
Dishmaker is a machine the size of your dishwasher, goes in your kitchen. Makes your dishes when you need them and then recycles them when you are done. Uses plastic discs

The Distant Gardener. What conversations in the telegarden reveal about the user experience of telepresence. Used chat data to understand telepresence. Telegarden is a robot in a garden that can be controlled using a web interface to plant seeds and watch them grow. The more users engaged in chat, the more likely they were to focus on technology outside of the garden.

Distuinguishing magnitude and frequency of vibrotactile effects with tactile mouse and trackball.

Vibrotactile feedback is used in everyday devices (cell phone), but is used in only a basic way; there are many potentials for this - the major problem is determining adequate feedack levels of encoding information. Vibro-tech was added to a mouse and a trackball and then tested.

Attention headphones: Augmenting conversational attention with a real world TiVo

Computer users in public environments require sociable ways to filter out noise generated by other people. There is an indreasing need to block noise; noise cancelling headsets are sensitive to social request for attention - you become socially unaware of what is going on around you.

Block noise, using eye gaze to establish an ad hoc social network. Route audio between headsets within the same social network, as determined by proximity and eye contact.
Allow noise - tivo for the real world, increasing attention capacity via audio buffering.

Drawing in Space, using the 3D Tractus (latin for drawing, space and movement).

goals are to augment the classic drawing board with a 3d tabletop interface. 3d curve input and surface models are captured.

Editble User Interfaces: Seamless Interfaces between People, Data and Food. monzy@stanford.edu

GUI to TUI to EUI; Beancounter is six hollow tubes filled with jelly beans. Computer actuated valves control two stage dispensing of flavored jelly beans. TasteScreen EUI sits on top of your LCD monitor with various flavoring atengs, dripping onto the monitor. Complexity of Flavor; flavor has four basic descriptors, but it leaves us at a loss if we are trying to distinguish between subtleties.

Designing EyeTap Digital Eyeglasses for Continuous Archival and Retrieval of Experiences. Steve Mann. capture and redisplay personal experiences exactly as you see them, but mediate vision to enhance the way you see them. Aremac (microdisplay), Camera, Diverter (optical element). http://www.eyetap.org

(at CHI05)

Accessibility

Web Accessibility for People with Cognitive Disabilities
Oregon Health and Science University
Richard Appleyard

Mostof the research to date in this field falls into two main categories: Special education and cognitive rehab (relearning of skills - uses high interaction excercises or games to do that). These are "drill and Practice". The other main area is in assistive technology - advanced, such as eye tracking and speech synthesis, and multimedia (assisted self-selection of career preferences).

W3C Web accessibility initiative, from 1997, "the power of the web is in its universality. Access by everyoneregardless of disability is an essential aspect" (Tim Berners-Lee). The guidelines gave priority to perceptual issues and less on cognitive issues.

Version 2.0 of the w3c spec will advance to include issues like "predictable interactivity" and "understandable content".

Four main cognitive components of web nagigation: situational awareness (momentary knowledge of awareness), spatial ability and memory (awareness of content relationships and navigation), task set switching (the ability to move quickly from one task to another), and Anticipated System Response.

Research question was: Can people with mild to moderate developmental cognitive disabilities navigate the web, and what are the barriers to accessing information on the web for people with mild to moderate DCD? (intellectual disability - mental retardation).

Study population was "convenient" (n=27). Diagnosed with mild to moderate mental retardation (IQ of 40-70), free of disruptive behaviors, adult, and able to participate in a meaningful discussion.

Sat them in front of a computer, asked them to think out loud performing tasks (basic: click on the sites header, highlight sentences on the page, find an interesting part of the web page).

Mixed-method approach: think aloud and "guided walkthrough". Majority of users encountered major usability errors; people with DCD do use the web but have significant problems even accessing WCAG-compliant web sites.

www.oidd.org for more information

Dance Along: Supporting positive social exchange and excercise for the elderly through dance.
Pedram Keyani.

Common perceptions of the elderly: people with limited mobility, declining health, and mental decline. However, this is a group with a wide range of physical and mental conditions, but have experienced a change in social structure. They encounter barriers to social exchange.

Goals are to provide entertainment and excercise for the individual and promote social exchange within the group. You don't stop dancing from growing old; you grow old from stopping to dance.

Dance Along is an augmented dancing environment. Projected scenes from famous movies, a dance floor, and a simple interface for song selection.

Iterative design process included literature review, a pool of ideas, interviews, augmented dancing, multiple rounds of interviews and user testing, and the final design.

Trends in elderly research include the importance of social networks, the difficulty in maintaining excercise, the cultural importance of dancing, and interactive music displays.

Interviews and observations showed that there were exclusive cliques and barriers to conversation and social exchange. Found that excercise is important to elders, and dancing is important to local elders; however, there is a lack of male partners and a lack of dancing events.

Iteration 1: Dancing to Footsteps. Projected footsteps to the dance scenes.

User Testing 1: Five people found that the footsteps were much too difficult to follow; the participants only faced the screen.

Iteration 2: Added extra screens in order to have more engagement with the audience. Also added a table with a scroll knob to select items on the screen.

User Testing 2: 100 people in attendance, lasting 1 hour. Simply having a dance floor in the celebration offered them a chance to get up to move around. Encouraged social interaction through movement.

Future Design Guidelines: provide a proxy for reminiscence, a range of support, and excercise through entertainment.

Human Computer Interfaces for Autism: Assessing the Influence of Task Assignment and Output Modalities
Ouriel Grynszpan
http://www.limsi.fr/Individu/ouriel

Autism is a pervasive developmental disorder. Three criteria: social interaction is impaired, communication is impaired, and restrictive and repetitive activities of interest. There are a high inter-individual diversity within people with autism; often linked to mental retardation, but does not have to be; "high functioning" autism have large IQ scores.

Psychopathologists' practice (Murray, 1997)
Experimental Studies have shown the usefulness of computers in education for Autism (Moore and Calvert, 2000)

But have not addressed design issues to help computer developers create educational HCI for autism.

Method: 8 teenagers with high functioning autism participated in a 13 session workshop (1 session a week). Two domains of learning were studies: social dialogue understanding and spatial planning. Expected people with autism to be more skilled in spatial planning than in social understanding. Also compared different interfaces with different modalities (speech, image and text) and compared before and after the training period.

People with autism have a hard time understanding abstract concepts showing irony or metaphor; they understand expressions literally.

"Excuse me, I need a hand".
"I don't want to give you any of my hands"

The goal of the game is to identify the socially awkward element; uses voice synthesis to read the text for those with reading problems.

Number of correct scenarios; significant results are that there are more correct scenarios after training than before in the dialogus training, but no difference in the spatial domain.

There is usefulness in computer assisted education for social skill training of people with autism; multiple simultaneous signals *may* increase the complexity for people with autism.

Designing Interactive Life Story Multimedia for a Family Affected by Alzheimer's Disease: A Case Study
Tira Cohene, University of Toronto

Communicating and interacting with participants is useful while conducting user centered design ethnography. There are obstacles, however, to having sufficient interaction with the participants. Language, cultures, etc.

Alzheimer's Disease is a degenerative brain disease, causes memory loss, cognitive loss, lack of language skills, and difficulty performing routine tasks. This causes problems interacting with end users.

Intervention technology is designed to enhance the well-being of people's lives; this may focus on support, etc. Life stories are an example of intervention that use personally historic information to support one's identity. The goal of these tasks are to support Rehab, but not to support social interaction.

Interactive Life Story Multimedia case study. Hypothesis is to support a positive life identity, and provide a stimulating experience. Collect life stories, explore reactions, and encourage and support interactions. The challanges, however, include the unique user group. Memory loss is just one of the symptoms they may have. They may also have trouble recalling words, making decisions, etc. Also, ethics are a large issue - not disrupting the lives of the users.

Performed ethnography, contextual inquiry, interviewing, participatory design, and prototyping. Performed this through moderated group activities, exploratory and receptive contextual inquiries, open minded and resourceful interviews, probes, and assorted story gathering.

Preliminary testing found that basic adaptations to user testing techniques helped in gathering data.

results: Needed to support tasks on a much lower level, by adding more prompts in the system. Created moderated, exploratory and receptive techniques, and needed to be more resourceful and foster imagination and inspiration. The implications of this on secondary stakeholders was (seemingly) positive; the family participated in the design process, and ultimately became a central focus in the design efforts.

Blind Learners Programming Through Audio
Department of CS, University of Chile
Jaime Sanchez

Audio based interactive environments intended to enhance cognition in blind children. Some of the learners want something more interesting than just environments; they want to learn to program (!).
Most programming language are focused on sighted learners, based on a "GUI" (a visual programming language). Blind learners then have to memorize command lines and write commands correctly.

Most blind programmers have to use text to speech to read commands and variables, assuming they follow the same logic during programming as those with sight.

So what are the particular needs of novice blind learners and why is it difficult to them to map and understand programming languages?

APL = Audio Programming Language for novice blind learners.

APL is an audio programming language for novice learners - a tool use to learn the basic programming concepts, to fit the needs and mental models of blind novice learners, and to motive learners to enter the programming world.

Results: APL did not support multiple cycles, or allow for waterfall conditions; also, the sound and voices were confusing.

Back from dinner with Lisa, Cliff, and the Savannah kids. Great meal; I'm totally exhausted. We went to an excellent little Italian restaurant named La Buca, and had a tremendously inexpensive and wonderful meal. Now, I will pass out.

(at CHI05)

Behaviour, Realism and Immersion in Games
Keven Cheng

Immersion in Games.
My Metal Gear Marathon. Discussed playing Metal Gear Solid II; ended up playing it for 17 straight hours. The console begins to "freak out"; the characters in the game begin talking. "Turn off the console". "You've been playing for a very long time; maybe you should take a break". Immersion is a powerful thing in a game, and may be something we are missing in other areas of design. How do you build this type of commitment?

Built off Emily Brown's presentation at CHI in 2004, Barriers to Levels of Immersion. Playing, Engrossed, and finally Presence. At the same time, you can invest time, effort, attention, and access.

There are barriers between levels; these are "game construction" and "empathy and atmosphere".

Within game construction, one needs certain elements to succeed.

There is some sort of connection between expectations and results. We build expectations based on visual styles as well as graphics. Realistic graphics lead to realistic behaviours (?).

If engrossed, does that break one out of immersion? In order to break one out of immersion, one first needs to be immersed.

Premise is to see what happens when immersion is broken due to extremely uncharacteristic behavior. Numerous participants never noted any change in behaviour throughout their play.

(at CHI05)

Attention-based Design of Augmented Reality Interfaces. MIT Media Lab.
Attention is a limited resource; in order for a system to be useful, without detracting, it needs to consider how the resource should be allocated. We need the simplest interfaces that we can' the principles include Exogenous/Endogenous visual cues, parallel/serial visual search, and object recognition. The interfaces need to be intuitive.

Show the user a cue in order to force a user to find a specific element; you can use gestalt principles to encourage visual cues. Spatially distributed visual cues.

Next level is cognitive. All of the elements in the kitchen are remote; how can you tell which burner is hot? We use intuitive mapping to understand which knobs drive which burners; project information on the burner (such as a red surface on the correct burner) in order to do away with remote controls.

Familiar mapping allows us to understand the termperature of the water. For example, cold water coming out of a tap is blue; hot water coming out of a tap is red. People are able to understand what these elements mean because of the nature mapping of the ideas.

Augmentation as aggregation of tools. Prototyped the kitchen to try to visualize their interface. For example, fridgecam allows you to stand in front of the fridge and see the items inside of it, without having to open it (? why on earth would you want to do that)

(at CHI05)

Wizard of Oz Interfaces for Mixed Reality Applications. Steven Dow.

DART: www.gvu.gatech.edu/DART

When we talk about Wizard of Oz, we think about the all powerful wizard that is really just a man behind the courtain twisting knobs and pulling levers. We use this in HCI in the same way; the human, behind the scenes, controls some aspect of the computer that will eventually be replaced by a human. Thus, WoZ facilitates the interaction between the human and the computer. This can be used throughout various stages of design, but is rarely used more than once throughout a design evolution. There seems to be a need to co-evolve the wizard interface throughout the process.

The problem addressed in the paper is that there is a lack of support within the design process for this technique. Designers Augmented Reality Toolkit, or DART, supports iteration.

DART promotes event-based broadcast communication; there are cues and actions. The cue is something the user inputs, and the action is some change to the system or output. The WoZ tools can be thought of as a Puppet of Oz + Wizard of Oz.

(at CHI05)

Magic Land: Live 3D Human Capture Mixed Reality Interactive System. MXR - Mixed Reality Lab, Singapore

Adrian Cheok

The roots of the project grew out of a glimpse of how people are communicating. In 2001 Space Odyssey, people communicated via videoconferencing; and that's exactly what happened. That future is now; Star Wars, however, viewed the future as a 3d interactive space, where the physicality and the virtual combines.

How can we bring human/human communication to the next level? (telephone => video conferencing => mobile phones => AR conference => 3D Live)

Magic Land. How can we merge the real and the virtual world? A cross-section between art and technology. Demonstration of the advances in HHI and HCI (3d live, human capture and interaction) .. mied reality merging, and tangible interaction, along with the artistic new approaches of delaing with live mixed reality content for artists of any discipline. Bring to attendants new kinds of art creation and reception.

The artistic aspect of this installation introduces to arists easy, tangible and intuitive approaches in dealing with mixed reality content.

(at CHI05)

Smart Laser-Tracking for Human-machine Interface
University of Tokyo

The problem studied involved how to create interaction on ever shrinking devices, without using the WIMP/Alphanumeric input and the touch sensitive screen/stylusstyle. Input space is merged with the viewing space; you can reuse the space. Perhaps, the next step is to remove any input output space at all, or remove the need for additional input devices. Create ergonomics-independent miniaturization.

How do you remove the input space? (using non-invasive techniques).

Provide the PDA with intelligent sight: hand/finger position, handwritten character recognition, gestural recognition. These have drawbacks; they may require intensive computation, may not be robust, and are sensitive to external issues (illumination). Passive-vision input is versatile but not mature yet.

Hand/finger is closer than anything else. The fingertip is always visible - there is no occlusion, and the fingertip shape is always simple and stable. A proximity sensor with some angular information will help us provide suficient data for tracking (same as how blind people find their ways, same as how animals find their ways).

Cost-efficient, robust and fast! But may be cumbersome and annoying.

The proposed antenna, then, may be simply a directive beam of light. This would use active lighting, provideing and controlling illumination. You can measure the surface roughness, speed, or even biometric data. The laser can also be used as an output device. There is no haptic feedback, however - there are no visual cues.

So how realistic is it to integrate this into a PDA? MOEMS (micro opto electro mechanical system) is an electronic circuit, and photodetector, and laser diode and mirror on the same ship.

Hardware uses discrete components (a laser diode, a pair of mirrors, a single non-imaging photodetector).

The tracking principle starts with a smart laser scanner. The laser excursion is locked around the areas of interest. Circular laser "saccade" is centered in the finger. Tracking sequence is repeated every millisecond.

Absolutely incredible; a gestural system with no physical intrusion on the user, allowing for a much more poetic experience.

(at CHI05)

Improving Automative Safety by Pairing Driver Emotion and Car Voice Emotion
Clifford Nass

Emotions and moods affect performance, attention and cognition. Positive emotions create a positive effect on problem solving and decision making.

Emotions are contagious, and interfaces can manifest emotion in graphics, linguistics and paralinguistic cues (speed of talking, tone of voice, etc)

This is important because driving performance is based on attention, judgement and attitude. Driving performance is also affected by mobile phones, passangers, in car information systems, and other things.

Thus, how do user emotion and car voice emotion interact to influence drivers' performance and attitude?

Experimental Design. 2x2 - Happy/Sad, Energetic/Subdued.

Used a driving simulator - a video game (Hot Pursuit).

Results:

Match groups (being happy and having a happy passanger, being sad and having a sad passanger) have better driving performance, more engagmeents, and better interactions.

Thus, we need to be aware of the users state in order to match it.

(at CHI05)

Private Communication in Public Meeting.
Nicole Yankelovich
Sun Microsystems, Inc.

Research team at Sun studies the improved effectiveness of distributed meetings. What were problems that people have with this type of meetings?

They conducted user resaerch, identifying the top problems people have with distributed meetings. Audio, behavioral, and technical problems are the largest set of problems encountered.

Additionally, the need for awkward behaviors is strange, socially; having to speak your name, or otherwise distrupt the flow of communication.

Wanted to develop user interfaces that specifically address these problems.

Didn't build any UI support to address roles, or things like that - these type of issues are socially mediated.

The Facilitator is the piece of software dealing with the audio portion of the meeting. Medium-deployment of the tool within SUN, and gathered feedback.

Making problems visible was helpful, but was not sufficient. For example, if there is a problem with audio quality, it is only uni-directional to flag the problem. Users had no way of qualifying the type of problem or discussing the problem in greater depth.

Usability study; six employees, four previous users and two new users. Found that comfort level with chat correlated to perceived legitimacy of uses. Users felt that the "whisper" command was not useful - because it was rude to whisper in a meeting!

Almost all of those who used voice chat whisper sessions involved only two users, and the majority were short.

How can they try to get "whispering" accepted as a legitimate thing to do during a meeting? Renamed it to "voice chat".

(at CHI05)

Ing-Marie Jonsson
Thank you I did not see that: In-car speech based information systems for older adults

As people get older, their perceptual, cognitive and motor skills degrade; they have trouble determining speed and distance. Vision narrows; the peripheral is degraded. The biggest risk factors for the elderly include maintaining a decent speed, making turns, heavy traffic, and new signs and signals (changes to routine).

Older adults benefit from reminders and timed information. In car information and hazard system has a positive effect on younger drivers, if the system is accurate.

Thus, the focus of the research is: Will a spech based in-car information system help or hinder older adults, and will the age of the voice influence driving performance and the perception of the in-car information system?

Driving performance. The # of accidents was low when driving with a helper program in a younger voice; middle when an older voice was present; large when there was no voice.

Clear connection between the usage of a "young voice" and "good results"; In car information system with hazard and traffic information can help older adult drivers. The age of voice for in-car information systems can influence perception of systems and driving performance of older adult drivers.

(at CHI05)

Wendy Kellogg, Shumin Zhai, papers chairs.

Introducing Randy Pausch. Professor of Computer Science, HCI, and Design at Carnegie Mellon. NSF Principal Investigator.

Cross Disciplinary Collaboration, Confessions of a technologist who has worked with psychologists, artists, designers and other creatures who are strange to me.

We must focus our energies on the right things. A metaphor for CHI when it is not doing a good job. We have finite human capitol, and we must spend it in the right places. Interdisciplinary collaboration is the right place to do that.

Why do we need creativity?

We are in a world going .. insane. Mcdonalds is offering cell phones with mcdonalds. We can build anything. Technology and resources make anything possible; now, we need to figure out WHAT to make.

We are living in the future. Star trek as a scientific predictor of what is to come.

Creativity comes from putting different kinds of people together. They see the worlds differently; its partly innate, and partly training. Having the most diverse team allows us to create a world where one sees the world differently.

You put blinders on; you see the world through a filter, and the filter is selected by the major you choose.

Diversity by gender; CMU CS majors are 35% female. Utility, not political correctness; Now that computer scientists are involved in designing things we use, we really don't want only one gender playing design.

How do we encourage creativity?

1. Diversity by point of view; we need to learn to see the world from as many viewpoints as possible. Creativity is more valuable than being smart! If you are in the top 1% of your class, there are 60 million people smarter than you.

2. Don't be afraid to be silly. You learn this from your parents; you learn it from your culture. Don't be afraid to look silly.

3. Being childlike and childish is critical to being creativity. The last time we felt that there were "no wrong answers" was when we were children. Smell the crayon. Smelling a crayon makes you.. six years old again. Smell is our best memory evoking sense.

4. Thinking out of the box, so to speak. Don't try to become an expert in other areas; learn enough to converse and trust your partner. You need to know the vocabulary - you need to be able to communicate.

1995 - "My Eyes Get Opened". Walt Disney Imagineering. An environment with artists and engineers working together to make a virtual reality ride. An artists is in the business of creating a perceptual phenomonon.

Can you dulicate the artist:engineering one to one ratio?

ETC; Entertainment Technology Center. Professional masters degree; technologists working small teams to make things. The keys to the ETC are freedom. There is no dean to report to. There is license to break the mold. The entire curriculum is project based. And its intense, fun, and student based.

Edutainment. Developed with the fire department of new york city; hazmat training. Networking simulation to train 1st responders to chemical spills and terrorist attacks. The most valuable thing firefighters learn is how to communicate with others.

Quasi. Interactive animatronic figure; repsonds to environment; designed, fabricated, and programmed. 14 weeks, 6 students.

Communication tools. Messyboard; Adam Fass. What is the lowest barrier to entry for sharing whiteboard tools?

www.messyboard.org/blah

Lightweight method of communication, for arists, is crucial.

Leadership by example. Where do students go?

- Electronic Arts
- Disney
- IMAX
- video game companies
- academia/edutainment
- PhD
- startups

85 students, with a focus on collaboration. How much can we teach you, in two years, about collaborating with other people?

So how do you do this?

Each side must be there for their own vested self interest, and not in a service of the other. Amortize your opinion over a minimum of sixth months, before creating a judgement - double that for artists :)

When putting student teams together, start with the very best; otherwise, people's worst fears are realized about "the other people". The first person over the wall has to be great ('really kick butt').

By understanding what motivates other types of people, you can get along better.

Rich Gold; talks about how we normally think of drawing/math .. right brain/left brain.

Instead, think of it .. Artists and Scientists both search for truth. Design and engineers search for problem solving.

Great training is a double-edged sword. The biggest problem designers have is that they design for the "old medium". Every problem cannot be solved with color, layout and technology. Interactivity is new, important, and needs to be taught first.

When you work with artists, everything is tied together; everything is personal. Artists follow intuition, and know in their guts what the rest of us can find via experiments at extreme costs.

When you work with psychologists, they have incredible insights into what is really causing a phenomenon and how to prove it. At their worst, however, they are obsessed with measuring things right rather than measuring the right things. The more concretely you can know a fact, the less useful it is to know it.

Engineers and computer scientists are helpful people, and problem solvers, but are arrogant! and are dilbert! We also have a lack of empathy.

Orbiting the Giant Hairball

Got the booth set up with small technical issues, but overall, went well and I'm a bit less intimidated with the other projects that I've seen (although technically strong, a bit lacking in aesthetics and emotional creativity - something Michael's product is quite strong in).

Now, waiting for the keynote; Randy Pausch from CMU! I'm excited to see what he has to say and how his thoughts have changed since I was in school.

Arrived, unpacked, and now trying to get the wireless internet in the hotel to behave.

I was walking back from the Convention center when a strange man approached me; he asked me if I had some quarters. I said yes, and jingled the change in my pocket.

He looked at me; there was a strange pause where I thought maybe he was going to mug me; he said "naw, man, I mean crystal meth".

I laughed. He walked away.

Portland is neat so far :)

The trip is off to a fairly good start so far; no flight issues, although I almost ended up next to an overly large and heavy-breathing woman who seemed more interested in adjusting in her seat than in anything else. I'm getting more patient with people, generally, but when jammed into a small space like sheep, one can get fairly territorial in a very short period of time. Get your damn fat on your own side!

We are about an hour and a half outside of Portland right now; Michael is playing his PSP, and there is some terrible Damon Wayne garbage on the mini televisions in the plane. No one seems to be watching, or listening, or purchasing $2 earphones. I like flying; it gives me an opportunity to critique both new people and new products, and it certainly puts my own small life in perspective.

When we get in, the plan is fairly un-planned; I want to check out the conference facility and perhaps set up Triba, but I'm not sure if it will be open at 7:30. Also, I have no clue how big the city is, and I'm supposed to meet up with a bunch of kids for drinks. I'm also getting exhausted. I'll play it by ear; that seems to be the best way to handle conferences, generally.

Well, I survived another birthday. I'm old, and starting to feel that way (at least compared to the students :D) ..

Off to CHI tomorrow, and that should be fun. I'm bringing a small group from SCAD, and I think Lisa and Susan are going to be there, as well as Kevin and Tom (but no Wilson). No clue about the kids from Trilogy, but it should be a good time - I'm excited to see Portland. I've always had a small desire to live there, with no real rationale behind it.

And that is that.