Jiajing is taking notes

Post notes of my courses, my work and other things.

[jiajing-notes]

Reciprocity Theory

I found there are a couple versions of reciprocity theory.

Social Psychology

One is Newcomb's from a perspective of social psychology, talking about interpersonal attractiveness.

Newcomb, T. (1979). Reciprocity of Interpersonal Attraction: A Nonconfirmation of a Plausible Hypothesis. Social Psychology Quarterly, 42(4), 299-306. Retrieved from http://www.jstor.org.proxy.library.cornell.edu/stable/3033801

Altman talks about revealing informaiton and feelings to each other.

Altman, I. (1973). Reciprocity of Interpersonal Exchange 1. Journal for the Theory of Social Behaviour, 3(2), 249-261.

CSCW

In Building a successful community, they present three studies that show newcomers who receive comments, feedback, even negative ratings engage with the community more. Even though they did not mention reciprocity theory, it can be explained in this way.

I started searching CSCW paper and this one has some theoretical explanation of reciprocity:

Qian Zhao, Zihong Huang, F. Maxwell Harper, Loren Terveen, and Joseph A. Konstan. 2016. Precision CrowdSourcing: Closing the Loop to Turn Information Consumers into Information Contributors. In Proceedings of the 19th ACM Conference on Computer-Supported Cooperative Work & Social Computing (CSCW '16). ACM, New York, NY, USA, 1615-1625. DOI: https://doi.org/10.1145/2818048.2819957

It cited a bunch of paper, such of Cosley's paper. It defines reciprocity:

Reciprocity: people make or withhold contributions to others based on how others treat them (Rabin 1993).

Economics

It also works in economics and game theory. This paper is cited by Cosley. It talks about some social phenomenon and proved by math. But the word reciprocity only appears once, in Reference.

Rabin, M. 1993. Incorporating fairness into game theory and economics. Am. Econ. Review 83(5):1281–1302.

So the paper in the reference of Rabin's is:

Goranson, R. E., & Berkowitz, L. (1966). Reciprocity and responsibility reactions to prior help. Journal of Personality and Social Psychology, 3(2), 227.

Zhao et al.(2016) cites Falk et al. (2006)'s paper:

People are reciprocal if they reward kind actions and punish unkind ones.

Their theory takes into account that "people evaluate the kindness ofan action not only by its consequences but also by its underlying intention".

Falk, A., & Fischbacher, U. (2006). A theory of reciprocity. Games and economic behavior, 54(2), 293-315.

They also cites Gouldner's work, reciprocity is analyzed as a social norm: the social norm of doing for others because they (either specifically or generally) have done or will do for you.

Gouldner, A. (1960). The Norm of Reciprocity: A Preliminary Statement. American Sociological Review, 25(2), 161-178. Retrieved from http://www.jstor.org/stable/2092623

[jiajing-notes]

wk 9 Breaking the rule of conversation

In our daily communication, there are some implicit conversation rules, e.g. you are supposed to answer questions if I ask you, I should greet the other if she says Hi to me.

But some violation of rules makes the question interesting.

Maxim of quantity

Imagine a scenario, your colleague asked you “How is your day?”, you said, “Oh I woke up at 7 this morning, I got some coffee first and then read some newspaper…” You know it is inappropriate because you are providing excessive information. Either too little or too much information broke the rule of conversation. It is the maxim of quantity.

Maxim of relevance

Here is an interesting paper that questions human’s ability to detect incoherence in conversation. Galantucci & Roberts (2014) did a study in which two pairs of participants chatted about two pictures. The images in each pair were the same except the color. Both of the images for two pairs had five famous figures but in different settings. As the picture below shows.

Each pair knew their cartoons were different in color but they did not know the existence of another pair. Over the course of the conversation, there were four 30-second crossing during which a random member of Pair was re-partnered to a random member of Pair B. After 15 minutes they were asked to answer questions about their feelings in the conversation and guess whether they were in the crossing or non-crossing group (50% in crossing group).

The result shows that over 40% of participants failed to notice that the conversation went wrong. The crossing detection rate was 57.9% and not significant. Although the study itself was problematic and had a various explanation, it is interesting to look at how people detect the relevance in conversation.  

Another study by Langer et al. (1978) tested how people process incoming information by using request, placebic info, and real info when asking for jumping into a line of printing. 

The result shows that when the number of copy is small (e.g. 5), the detection rates of placebic and real information were nearly the same. Only when you ask for 20 copies, people would think about it and may refuse. 

Are human beings really not good at detecting irrelevance? What about people’s performance in a non-chill situation? 

Rogers and Norton (2011) did some interesting studies about dodging. What if we answer the wrong question in the right way? which is a common strategy of politicians. 

They conducted a series of studies. In the first one, they asked each participant to watch a 4-min clip of a mock political debate designed to simulate an actual televised debate. Participants were randomly assigned to one condition of a 3 (goal: none, social, dodge detection)  x 3 (dodge: correct, similar, dissimilar) between-subjects design. 

The correct condition watched matched question and answer while the dodge one watched dismatched. The social goal means social evaluation such whether like this person or not. Dodge detection goal asked participants to pay attention to the question. 

The result is amazing. How low the detection rate is! 

Maxim of Manner

The subsequent study also revealed human’s failure in detecting dodging. Three conditions: correct answer delivered fluently, similar answer delivered fluently, and correct answer delivered disfluently were tested. 

It is good to see the recall rate of the question is sane. But interpersonal ratings (how much they trusted him, how much they liked him, how honest he was, and how capable he was) had no relationship with whether the answer was correct or not. So a good lesson to know, next time of your public speaking, if someone asked a question you cannot answer at the spot, you should say, “This is a good question, I do not totally understand your question, but here is my explanation of a similar one...” Then the audience will forget the original question and still like you!  

To summarize, there are three types of conversation rules:

Maxim of quantity

Maxim of quality

Maxim of relevance

Even though this week’s readings were not computational papers, they are so insightful and interesting. 

Reference

Galantucci, Bruno and Gareth Roberts. 2014. Do we notice when communication goes awry? An investigation of people's sensitivity to coherence in spontaneous conversation. PLoS ONE 9(7).

Langer, Ellen J, Arthur Blank, and Benzion Chanowitz.1978. The mindlessness of ostensibly thoughtful action: The role of "placebic" information in interpersonal interaction. J Pers Soc Psychol 36 (6): 635.

Rogers, Todd and Michael I. Norton. 2011. The artful dodger: Answering the wrong question the right way. Journal of Experimental Psychology: Applied 17 (2). 

[jiajing-notes]

DIS Paper Using Interviews as Research Methods

This post is about methods and writing.

This week I read some DIS (Designing Interactive Systems) papers which use interivews as research methods. Here I will give some examples and summary of the reading.

The challenge of these interview projects for DIS is that they do not DESIGN anything. Instead, they do plenty of user research as a preliminary step to design. If we just talk about the research from the perspectives of social science or communication without any implications for design, the paper may not be appropriate for DIS.

Emphasize of the goal of the paper as design

“More than just Space”: Designing to Support Assemblage in Virtual Creative Hubs (Luik et al., 2018)

This paper aims to understand interactions at creative hubs, and how this understanding can be used to inform the design of virtual creative hubs. They make it clear in the keywords: informing design and they repeatitvely mention the goad as "inform the design of virtual hubs" for several times in the paper.

We believe that our findings and their analysis through the lens of assemblage will benefit the field of Human- Computer Interaction (HCI) by informing the design of social-technical infrastructures that seek to ...

Our aspiration in conducting this research was to inform the design of interactive technologies that...

Caller Needs and Reactions to 9-1-1 Video Calling for Emergencies

This paper explores how video calling services should be designed through an interview study with people who have called 911 in the past. This paper is merely an interview project. It focuses on the challenge of 911 calling systems and provides implications for future design.

Our overarching goal is to understand how to design ... systems from the perspective of xxx, with an emphasis on matching the technology with xxx needs.

These issues illustrate that designing xxx requires careful design considerations to balance the needs of xxx and xxx.

List the contributions to design

“Hey Alexa, What’s Up?”: Studies of In-Home Conversational Agent Usage

This paper used mixed methods - survey and interviews to explore how people interact with Alexa:

Resulting knowledge will give researchers new avenues of research and interaction designers new insights into improving their systems.

Our contributions are as follows... Additionally, we provide information about the limitations and shortcomings of the technology as it is designed today, and suggestions for how CUIs can be better designed in the future.

Demostrate more design elements

Evaluating and Informing the Design of Chatbots (Jain et al., 2018)

This paper emphsizes more on implications for design.

Here is what they write in the abstract:

We conclude with implica- tions to evolve the design of chatbots, such as: clarify chatbot capabilities, sustain conversation context, handle dialog fail- ures, and end conversations gracefully.

They demonstrate and discuss some interfaces of Messenger app. In the Discussion part, they offer design implications for both chatbot designers and chatbot platform UX designers, as well as potential future usage of chatbots in HCI research. Although the implications are a bit trivial and confined in Facebook Messenger which looks like a usability test, this paper gives more sense of what a design paper would look like (the authors are industry so it quite makes sense).

Examining Self-Tracking by People with Migraine: Goals, Needs, and Opportunities in a Chronic Health Condition

This paper also spent much space to talk about design implications in Discussion section.

Create concept to demonstrate the idea

This kind of paper puts forward a new concept and demonstrate the design of technologies to support it.

Grounding Interactive Machine Learning Tool Design in How Non-Experts Actually Build Models

This paper is different because it not only shows the results of interview, also offers a concept/ framework for future design (the first author herself is a designer). Although it is not a detailed design work with evaluation, the concept has offered enough implications for future work.

Investigating How Experienced UX Designers Effectively Work with Machine Learning

This paper has the same author with the one above. It probes several open questions at the end of the paper, which is also creative.

Social Support Mosaic: Understanding Mental Health Management Practice on College Campus

This paper investigates how emerging adults seek help managing their mental wellness through social support. They define the support network as "the Mosaic of Scoial Support" and discuss the technologies to support it.

Reference

I am too lazy to write this part... will add it in the future.

[jiajing-notes]

CMC - Informal Communication

According to authors' on-site study and the analysis of previous data, informal communication help to increase familiarity and mutual respect, the likeliness of collaborators, which raises the chance of the success in collaboration. To create the chance of informal communication, two systems were designed and analyzed in this paper, they introduced two examples: VideoWindow and Cruiser.

VideoWindow connects two public areas with a life-size screen. People from two locations can talk and interact through the "mirror" in the hallway. This system looks cool but still have several design flaws. First, the usability of the system has some issues. The position of the camera and microphone failed to provide smooth high-quality video conversation. People either find it hard to hear others or move out of the camera range. Second, strangers do not benefit from this system if they do not meet frequently. In other words, VideoWindow only works for two groups of people who already know each other.

Cruiser is a switched telecommunication system that allows users to "cruise" in a virtual hallway. Users can switch video from room to room until he/she finds a person to talk to.

There are two primary issues in technologies of informal communication. First is users' ability to control their privacy. In a shared physical situation, two people can lower their voice or move to the corner, but in VideoWindow it is hard to keep the privacy. Second is the interruption of others' work in Cruiser system because you cannot knock the door as you do it in the real world. The author gave us some implications such as selective media blocking, reciprocal views, visual status cues, and user-specific filters.

As the author said, the value of informal communication is asymmetrical. We made a great endeavor to create chances for such a common and easy communication, which may not work in the end.

This paper was in the 1990s, currently there are other solutions for informal communication, such as Beam. It is more acceptable for people to come into a moving robot in the hallways without having the feeling of being mintored or interrupted.

Reference

Kraut, R. E., Fish, R.S., Root, R.W., & Chalfonte, B.L. (1990). Informal communication in organizations: Form, function, and technology. In S. Oskamp & S. Spacapan (Eds). Human reactions to technology: The Claremont Symposium on applied social psychology (pp. 145-199). Beverly Hills, CA: Sage.

The Nature of Informal Communication The Dynamics of Informal Communication in R&D

Episode 1: The call

Episode 2: Stop Short

Episode 3: At the Candy Machine

Shared Features of Episodes

Surface Characteristics of Informal Communications

Frequency and initiation

Duration of Informal Conversations

Summary

The Content and Uses of Informal Conversation

Preceived Value of Informal Communication for Production and Social Functions

The Effects of Informal Communication on Person Perception

Informal Communication and Momentum in Collaboration

Informal Communication and the Success of Collaboration

Indices of proximity

Familiarity

Liking

Summary

Process

Outcomes

Other Communication Modes

Collaboration

Physical Proximity

Organizational Proximity

Research Similarity

Summary

Technology for Informal Communication

Co-presence

Low Personal Costs

Visual Channel

The VideoWindow Teleconference System

Cruiser

Issues in the Trade-Off of Access and Control

Conclusions

[jiajing-notes]

Wk6 Quantifying and Calculating

This week we went through three articles in the theme of quantifying and calculating. These days we are clear about two different research methods: qualitative and quantitative methods. When you are saying “I’m a qualitative researcher.” There is some value-driven information here. In the world of quantitative research, how numbers work in a holistic way instead of being so polarized?

Espeland and Stevens - A Sociology of Quantification

This article addresses the oversight by analyzing quantification – the production and communication of numbers – as a general sociological phenomenon. It also introduces five sociological dimensions of quantification and call for an ethics of numbers.

Work. The work of collecting data draws our attention to where data comes from.

Reactivity

Discipline.

Authority

Aesthetic. Data visualization. People may have different training in interpreting the graphics.

Measurement intervenes in the social worlds it depicts. Cause people to think and act differently. It changes the dynamics.

Transform individual experiences and events into general categories or characteristics.

Alter relations of power by affecting how resources are distributed.

Create or reinforce categories to conceive human beings.

Then those measures began to control people. Seeing sth is the first step to control it.

Measures that initially designed to describe behavior can be used to judge and control.

Why and when people give authority to numbers, find numbers credible ways of knowing and communicating.

The authority of the information parallels the authority of its handlers in the hierarchy.

The authors are not talking about good or sad. They just suggest that numbers cannot be reduced to a narrow instrumentality but should be its ethics.

Hacking, “Biopower and the Avalanche of Printed Numbers”

Hacking’s this article is pretty interesting, which goes through the history of counting by isolating a certain period 1820s-1840s. At first census was just simple counting, later people created categories and people were fit into a list of categories. appeared and people were conceived into There are two revolutionary history about biopolitics.

A long term revolution was from the counting of sickness of British men to find the law of sickness. Long-term account biopolitics does not have subversive effect because it has little to do with new categories, new ways of conceiving people. There are two factors in the invention of categories of complaints.

One is the fetishism of counting, which brings easily applied categories in terms of to count. For example, classifying people according to their occupations such as “farmers, artists, salaried workers … ”

One is the theoretical and practical reasoning of individual professionals such as doctors. For example, there is an nationally standardized categories of mental disorders which requires doctors to categorize these disorders according to the list. It is ridiculous when you are illegal to die because the reason of death cannot be categorized into the list!

Hacking uses historical methods to draw how the data was collected. His point is to ask people to think about the conceptual schema people created. Some qualitative research method also uses quantitative way to deal with the data. For example, gender is qualitative information but we use 1 or 2 to represent them.

Bouk, How Our Days Became Numbered: Risk and the Rise of the Statistical Individual [Chapter 2]

Life insurance is not fair for African American. They were not offered life insurance, later, even though they had life insurance, they were only offered 2/3 of the benefits. The insurance companies did discriminate, but they did not discriminate against the race but the high rates of high morality. Race just happened to be a cost-effective marker of high morality. Here is a flaw in this prediction model that is the previous trend of morality rate in a certain period could not represent the morality right now. Since the black were no longer slaved, their life span could change also. Bouk’s viewpoint is not only deny the current prediction model like “you’re wrong with the source of data.” but the success of actual reasoning: “If you did it right, it would be fine.”

Reference

Bouk, D. (2015). How our days became numbered: Risk and the rise of the statistical individual. University of Chicago Press.

Espeland, W. N., & Stevens, M. L. (2008). A sociology of quantification. European Journal of Sociology/Archives Européennes de Sociologie, 49(3), 401-436.

Hacking, I. (2015). Biopower and the avalanche of printed numbers. Biopower: Foucault and Beyond, 65-81.

[jiajing-notes]

Wk3 - Rules and Regulation

This week we went through three articles about rules and regulation. One is Latour's ‘Where Are the Missing Masses? The Sociology of a FewMundane Artifacts. He talked about the relationship between society and technology. Another is Lessig's Code V2 chapter 7. I will not go through this article here. The last one is Radin's Regulation by contract, regulation by machine. She argues the illegitimacy of Digital Rights Management Systems (DRMS).

I do not gusrantee the accuracy of the content.

Latour - Where Are the Missing Masses?

Latour's this article is quite interesting. I like his humorous language like "you sociologist" and vivid example like the groom, map, and seat belt.

Latour wrote 10 pages about grooms and seat belt. These mundane tracing of detailed stories of human and technology gives us room to see closer about the relationship of society and technology, not only from a broad and structural view.

Here comes the argument:

Will moral vocabulary diminish because of the delegation? Is delegation negative?

Radin - Regulation by contract, regulation by machine

Radin discussed the contracts and the law for intelletucal property. Contract should be a private agreement, backed by the public law. When a conflict happens, people will turn to the state government. However, shrink wrappers, click wrappers, or browser wrappers ask to waive many rights to use their products. Radin's point in this article is that these so-called "contracts" are no longer legal documents on the court.

The law for intellecutal property is to grant someone the projection their knowledge. It is to give incentives to create instead of exclusive rights and prevent someone from using the products.

For example, a lot of developers and researchers scrapped data from websites (including me). It may violate Terms of Service (TOS). Should we make websites unscrappable, or create stronger law to project content on the websites from technical aspect (if it is possible)?

Reference

Latour, B. (1992). 10 ‘‘Where Are the Missing Masses? The Sociology of a FewMundane Artifacts’’.

Lessig, L. (2009). Code: And other laws of cyberspace. ReadHowYouWant. com.

Radin, M. J. (2004). Regulation by contract, regulation by machine. Journal of Institutional and Theoretical Economics JITE, 160(1), 142-156.

[jiajing-notes]

Summary - The Three Paradigms of HCI

This article is the summary of CHI'07 paper The Three Paradigms of HCI

What is paradigm

Before we look into the three paradigms, first we need to understand what paradigm is.

In Kuhn's theory of the structure of scientific revolution, science does not progress from accumulation of facts. Paradigms are successive overlapping waves in which ideas are fundamentally reframed.

In a paradigm there should be 4 elements:

What is being studied

The questions that are useful to ask about phenomena

How we structure our approach to answering the questions

How to interpret results

The authors argue that because:

HCI is in lack of reproducible experiments

Agres' generative theory: technical fields are structured around metaphors. These metaphors bring certain elements into the center and marginalise others.

The athors suggest to track metaphors of interaction to reveal the evolution in the three paradigms.

Three paradigms

Paradigm 1 (Industrial engineering/Ergonomics): Man-machine coupling

From industrial revolustion.

Identify and reduce human error by "optimizing the fit" between man and machine. It is objective in nature and driven by quantitative information.

Munition manufacturing, cockpit

Paradigm 2 (Cognitive Science/ CS): Information processing

Bridge the gap between machine processor and the mental processor

Regard both computer and human brain as information processors.

Windows, GUI

Paradigm 3 (User experience design/Anthropology)

Huamnize interactions by embracing the plurality of users and perspectives. Humans are perpetually embodied.

Siri

The 3rd paradigm

What is marginalised?

In the former two paradigms, several issues become marginalized in HCI (contentious claim):

participatory design

ethnomethodology

user experience design

interaction analysis

value-sensitive design

critical design

The authors see these issues as elements in the 3rd paradigm.

Principles of the 3rd paradigm

Intellectual commiments

The construction of meaning

Putting users in their place

Putting interface in their place

Putting researchers in their place

Explicit focus on values in design

The necessity, but inadequacy of theory

The underlying role of embodiment

Embodied interaction is

A shared intellectual commitment

A cross-cutting perspective at the heart of other commitment

The challenges of 3rd paradigm

Different ways of knowing - what it means to know sth is true

Sometimes the paradigms clash becaue they share different notions of what the knowledge is and how it is generalized.

Objective VS subjective knowledge

Generalized VS situated knowledge

Info VS interpretation

Clean VS Messy formalisms

[jiajing-notes]

Non-verbal Behavior

This seminar is about a highly influential paper about non-verbal behavior written by Paul Ekman and Wallace V. Friesen: The Repertoire of Nonverbal Behavior: Categories, Origins, Usage, and Coding

I have not read the paper yet, here I just post the content of the seminar first. 

Interpret non-verbal behavior from THREE aspects

Origin: Where is come from? How he chose this? It brings a great debate. Is it inherited? Or shared by humans? Is it cultural or individual? Do we learn it from the group?

Usage (the most the author discussed): What circumstance we use it? - External condition. Do it with others or do it alone? Victory sign

Coding: What info is under this gesture? How you delivery information?

Relationship to verbal. Does the gesture come before/ after you say sth?

Awareness. Are you aware of what you did?

Intentional. Do you want to do it or not? Note intention and awareness are different. You can do sth unintentional but aware.

Information conveyed. It is the most complicated. Do others understand?

Communication: what you Transmit. You want to have sth communicated.

Interactive: What you received. What is the reaction of others?

Extrinsic. What Outside the gesture - Arbitrary. It has nothing to do with what the gesture would be like. E.g. thumb - Iconic. E.g. victory sign - Pictorial. - Spatial. e.g. Point to the café. - Rhythmic. e.g. Do a certain gesture again and again. - Kinetic. e.g. Go to the door and touch the handle (Physical movement)

Intrinsic: What from gesture to the info

FIVE types of nonverbal behavior

Emblems. e.g. Thumb, you can use it instead of speak.

Illustrators.

Affect displays

Regulators. E.g. Nodding

Adaptors. Playing with hair/rubber band. Tap sb on the shoulder.

Batons. E.g. Trump likes to use fingers in his talk.

Ideographs. In the first sentence, you say New York is here, LA is there. Then in the later speak, you can keep using the two position because you already made an impression of the location.

Deictic.

Spatial. E.g. My son is this tall

Kinetogrpahs. E.g. I’m swimming

Pictographs. E.g. the gesture of shooting.

Human has the same response to smile, bad news and feeling happy.

[jiajing-notes]

Live Streaming in the Commercial Environment

THE LIVE STREAMING VIDEO REPORT: Forecasts, emerging players, and key trends for brands’ and publishers’ next big opportunity

The Live Streaming Video Report - BI Intelligent

Live streaming video refers to broadcasts in real time to an audience over the internet.

According to the definition, it should be equal to social TV.

person to person sharing

business to customer

business to business communication

Keypoint:

Live streaming video will further accelerate streaming videos overall share of internet traffic.

Live video’s value comes from its unique ability to add an authentic human element to digital communications. As a result, brands are leveraging three main streaming methods to connect with their viewers: tutorials, product launches, and exclusive and behind-the-scene footage.

Advertisers will continue to invest heavily in live streams.

The success of live streaming video hinges on brands overcoming a lack of measurement standards in the space, as well as changes in social media sites' algorithms that affect what content users see.

62 Must-Know Live Video Streaming Statistics - Livestream

81% of internet and mobile audiences watched more live video in 2016 than in 2015.

Twitter’s live streaming strategy takes form - Business Insider

The Facebook Live Streaming Benchmark Report 2017 - WOWZA

They surveyed over 700 video-streaming professionals across various industries in June 2017.

28%: in education, non-profit, government

26%: media

11%: streaming service provider

10%: live events producer

...

Summary

FB dominates the market

Broadcasters report that live streaming increases awareness and customer loyalty

Content varies

Major challenge: Setting up and integrating equipment and systems

Connetion/playback errors widespread

Statistics

82% FB Live is a strategic initiative (establishes trust and authenticity for the organization)

66% stream to FB live at least weekly

FB live -> multi-purpose, leads in event streaming

70% say their goal is to reach more audience

64% say a goal of connetion - deepen customer engagement

Benefit: flexibility + cost-saving

Magid's report about live stream

Facebook Live, YouTube Live Battle For Live Streaming Dominance - MediaPost (Research by Magid)

Market share

Facebook Live - 45% market share

YouTube Live with 44%

Instagram came in at No. 3 with 28% market share

Twitter, 19%

Snapchat, 17%

Twitch owned by Amazon, 12%

Livestream, 10%

Periscope, 9%

YouNow, 6%

UStream, 4%

Content

Breaking news - 24%.

Videos by friends or family at 23%

how-to videos, 21%

videos by YouTube personalities, 21%

comedic video, 18%

[jiajing-notes]

PaperReview: The cost structure of sensemaking

What is sensemaking?

Sensemaking is the process of searching for a representation and encoding data in that representation to answer task-specific questions.

New in this paper:

The emphasis on representation use and shifts in analyzing task performance. Representation: user interface

Identification of the learning loop.

Anytime principle: how tradeoffs can be made in one place for gains in other another.

Characterization of sensemaking

Case: laster printer

Sensemaker: course designer

The goal of the case is to design a course for Xerox technicians to understand the smilarities and differences of printers.

Main cost: data extration

Main gain: automated clustering.

For large investments, computer-based method is more cost-efficient than manual method.

Learning loop complex

Search for good representations -- generating loop, build a model

Instantiate representatoins (create encodons/data) -- data caterage loop, data collecting Representational shift: modify/revise the model according to the data.

[jiajing-notes]

Supporting Reflective Public Thought with ConsiderIt

ConsiderIt: a novel platform for support public deliberation on diffult decisions. Initially, it aims to support a community-written voters guide for Washington State in the contentious 2010 U.S. election. This paper they reported the deployment in 2010 as The Living Voters Guide (LVG), which attracted thousands of users and contributors leading to the election. In this paper, they present a new interactive approach for supporting public deliberation and describe a design strategy.

Paper Structure

Theory and design public deliberation

System description

Deliberation through position crafting

Exploring the guide to public thought

Deployment details and data

activity traces

user study: 7 participants never heard about LVG were recruited to help researchers get a sense of perceived value of LVG, learn how users understood and valued the interaction, how they reacted to points others submitted.

survey: short post election survey.

Results

ConsiderIt does not provide group cues to activate political identity.

The stronger a stance, the less likely they are to include a point written by someone tak- ing an opposite stance.

Discussion

Reference:

Kriplean et al. (2012). Supporting Reflective Public Thought with ConsiderIt. CSCW 2012.

[jiajing-notes]

Paper Review: Wikum: Bridging Discussion Forums and Wikis Using Recursive

Evaluation

two studies to evaluate the process of creating and reading a summary tree.

Study 1: Summarization

three discussion topics

two conditions: Wikum; Google doc with raw discussion text

Results

Faster in Wikum

Study 2: Reading and exploration

2-factor user study three conditions: Wikum, Google doc with summary, Google doc without summary. Procedure: participants read about 10 min. Then presented a list of 12 points and asked to select points they remembered. 6 of points were in the discussion while the other 6 were not.

Results

more accurate in Wikum than Google doc

Discussion

Design implications

A tension betweeen summary goals

Who summarizes?

moderators

role of original commentors

competing summary

crowd workers

Reference

Amy X. Zhang, Lea Verou, and David Karger. CSCW 2017. Wikum: Bridging Discussion Forums and Wikis Using Recursive Summarization.

[jiajing-notes]

The origins of the concept mapping tool and the continuing evolution of the tool

Paper structure

Introduction: history, issues addressed in the program, new tech enhances capabilities

The invention of concept mapping-history

Development of a new epistemology

Development of computer-based concept mapping tools

A new model for education

Conclusions

Piaget's theory: children could not profit in abstract concepts before they reached the formal operational stage of thinking, like age 11.

Fundamental questions in this program:

cognitive limitation - the result of brain development, or artifact of school or society?

Can 6 to 8 kids develop sufficient understanding to influence later learning?

findings in longitudinal study support Ausubel's assimilation theory?

Reference

Novak, J.D. & Canas, A.J. (2006). The origins of the concept mapping tool and the continuing evolution of the tool. Information Visualization, 5, 175-184.

[jiajing-notes]

Some thoughts about HRI:

Why robots are so popular these days?

We need the thing, or the human behind it?

We need a hug, or a person who cares about you?

If we have a remote partner, does it have to be a human?

What makes human-human interaction so special?

[jiajing-notes]

Summary: Research through design as a method for interaction design research in HCI

Gap

Design has much less impact in HCI community. The definition related to “design” is quite vague.

Method

literature review

workshop

semi-structured interviews: nine academic researchers, one graduate student, six interaction designers.

construction of a new model, iterative evaluations: presentations and discussions.

Result

A new model, following Christopher Frayling’s research through design: focus on making the right thing; artifacts intended to transform the world from the current state to a preferred state.

Contribution

allow the community to engage with wicked problems can’t be addressed through science and engineering methods.

feedback tech opportunity

a new method, transfer knowledge from research to practice.

allow interaction designer to make the contribution

motivate discussion

Content

Interaction designer use

true knowledge (models and theory from behaviour science)

how knowledge (tech opportunities from engineers)

ground exploration in real knowledge produced by anthropologists

to solve wicked problems, in which the stakeholders have conflicting goals.

The final output:

problem framing

articulation of the preferred state

artifacts, models, prototypes, products

documentation of design process.

Nigel Cross’s idea: design knowledge resides in the product, the artifact reflects a specific framing of the problem. (But he didn’t consider desgin practice to be regarded as research contributions.)

This model departs from the role of design researcher in LR:

member of practice team, doing upfront project research

study design process, analyze artifacts, discover patterns

critic

framing and communicating consultant

An obvious criticism: distinguish the contributions of design researchers as research not as practice.

to produce knowledge

research contributions should be artifacts that demonstrate significant invention.

ignore or deemphasize perspectives in framing the problem

researchers focus on making the right things, practitioners focus on making commercially successful things.

novel

demonstrate a significant advance through the integration.

not as agressive as tech in hardware and software, like inventing new product categories VS redesigning office furniture.

Their method unleashes interaction designers in a research context, to study the world, bring change, and avoid the means of justifying their researhch contribution from other disciplines.

What is research artifact?

Criteria for evaluating interaction design research

Process

Invention

Relevance

Extensibility

no expectation of same result

must provide details to make the research reproducable

or rationale for the selection of mehtod

constitute a significant invention

instead of validity, the benchmark for interaction design research -> relevance.

shift: what is true (behavioral science) –> what is real (anthropologists)

should have motivation, articulate the detail on current situation, and preferred state; otherwise the research appears to be self-indulgent and personal exploration but makes no promise to impact the world.

the ability to build on the resulting outcomes of the interaction design research.

described and documented in a way that community can leverage the knowledge.

Reference

John Zimmerman, Jodi Forlizzi, and Shelley Evenson. 2007. Research through design as a method for interaction design research in HCI. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ‘07). ACM, New York, NY, USA, 493-502. DOI: https://doi.org/10.1145/1240624.1240704

[jiajing-notes]

Why Condorcet Jury Theorem is faulty in collective decision making?

This is a summary of the article by David Austin-smith and Jeffrey S.Banks, Information Aggregation, Rationality, and the Condorcet Jury Theorem (see reference at the bottom). We were encouraged to read it before Monday's course. Of course I had a painful time reading it and the summary may not accurately convey the article's intension.

What is Condorcet Jury theorem?

The Condorcet Jury theorem states that under certain conditions, the probability that a collective chooses the correct alternative by majority votes exceeds the probability that any consitituent meamber of the collective would unilaterally choose that alternative. That is to say, majorities are more likely to select the "better" alternatives under uncertainty.

What is the problem?

This theroem has an important but largely implicit assumption that an individual behaves in exactly the same manner as when that individual alone selects the outcome. Actually, sincere behavior is not rational even has a common preference. And sincere voting does not constitute a Nash equilibrium.

You may ask, what is sincere voting? OK, there are three sorts of voting behaviors:

Sincere voting: each individual selects the alternative yielding his or her hightest expected payoff conditional on their signal.

Informative voting: each individual votes for A if and only if receiving a signal s = 0, and B when s = 1.

Rational voting: individual's decision rultes constitute a Nash equilibrium.

Attention: these three voting behaviors are not mutually exclusive.

The authors describe two assumptions:

Sincere voting is informative when one receives a signal that makes him/her think A/B is the true state.

The common prior belief that the true state is A is sufficiently strong that there is only one condition that someone will select B that he/she observes all three individuals' signals as B. (Why they talk about prior belief here? Otherwise how can you use Bayes' method!)

How to prove the claim?

The authors looked at the role the "sincerity" assumption in Jury Theroems with three variations of an extremely simple model.

(I will not introduce the model here because I have to admit I totally can't understand the definition and formula. After all they use the model to explain it and it seems quite interesting. I think I need to learn more about game theory to understand it).

Reference

Austen-Smith, D., & Banks, J. (1996). Information Aggregation, Rationality, and the Condorcet Jury Theorem. The American Political Science Review, 90(1), 34-45. doi:10.2307/2082796