Friday, April 22, 2011

Full Book Blog: Living With Complexity

Summary
The first chapter serves as an introduction to complexity. Norman differentiates between things being complex (state of the world, the tasks we do and the tools we use) and the idea of things being complicated (the psychological state of a person attempting to understand, use or interact with something in the world). He discusses keys to coping with complexity, he brings up how some complicated systems occur as a result of poor design and he provides several examples of how we like complexity in certain instances (like when we already know a lot about the topic or where complexity truly seems appropriate). He also mentions that one way to measure complexity is by the amount of time it takes to learn and master a task (10,000 hours).

In chapter 2 Norman brings up the idea of the conceptual model (the underlying belief structure held by a person about how something works), and how it helps to simplify the complexities of different systems. Norman discusses “featuritis,” adding more and more features, and explains how complexity is in the mind of the beholder. He also discuses how simpler looking does not always equate to simpler to use and how users tend to prefer an intermediate level of complexity.


mitpress.mit.edu
In chapter 3 Norman discusses how things get more complicated as the number of items increases. He provides remembering passwords as an example, and explains how we cope by putting the information in the world (pasting the passwords to the monitor or under the keyboard). He discusses the “scaling problem,” something that works well with only a few cases, but often fails as the number grows. He mentions how in an ideal world we would not need signs and how too much information makes things complicated. He concludes by discussing the benefits of forcing functions and provides examples through a discussion on toilet paper.

Chapter 4 focuses on social signifiers, indicators in the environment that allow people to navigate in otherwise complex and confusing environments. These signifiers are referred to as “perceived affordances” by some. Norman also discusses culture as another source of complexity and uses salt and pepper shakers as an example. Norman writes about the importance of signifiers and how they provide valuable cues as to the nature of the world and how people should act.

Discussion
Norman provides a lot of good examples throughout all the chapters that we read. He really digs down deep when explaining what complexity is, what causes it and the different types of complexity. I think one of the funniest to read was the one on toilet paper since he was so serious about it. I also thought it was interesting to read about how it takes 10,000 hours of work to be an expert.

Something I thought of while reading chapter 3 was how once when I was driving in a construction area, there was a sign on the left side of the road that read, “Left lane closed,” but on the right side of the road read a sign, “Right lane closed.” There were only two lanes and the cars were bumper-to-bumper. While I’m sure there would still be traffic regardless, I think things would have ran more smoothly if the signs had been accurate. Only the people who put the signs up knew which one was correct. (I think the left lane was the lane that was truly closed.)

Book Reading #52: Living With Complexity

Chapter 3
Summary
In chapter 3 Norman discusses how things get more complicated as the number of items increases. He provides remembering passwords as an example, and explains how we cope by putting the information in the world (pasting the passwords to the monitor or under the keyboard). He discusses the “scaling problem,” something that works well with only a few cases, but often fails as the number grows. He mentions how in an ideal world we would not need signs and how too much information makes things complicated. He concludes by discussing the benefits of forcing functions and provides examples through a discussion on toilet paper.

Discussion
Something I thought of while reading this chapter was how once when I was driving in a construction area, there was a sign on the left side of the road that read, “Left lane closed,” but on the right side of the road read a sign, “Right lane closed.” There were only two lanes and the cars were bumper-to-bumper. While I’m sure there would still be traffic regardless, I think things would have ran more smoothly if the signs had been accurate. Only the people who put the signs up knew which one was correct. (I think the left lane was the lane that was truly closed.)

Chapter 4
Summary
Chapter 4 focuses on social signifiers, indicators in the environment that allow people to navigate in otherwise complex and confusing environments. These signifiers are referred to as “perceived affordances” by some. Norman also discusses culture as another source of complexity and uses salt and pepper shakers as an example. Norman writes about the importance of signifiers and how they provide valuable cues as to the nature of the world and how people should act.

Discussion
In discussing signifiers, Norman mentions traffic and how people tend to interpret it as there being an accident up ahead when sometimes there is not. This happens so much in Houston (and likely in all other major cities). Whether it be a fire or an accident on another part of the highway, people tend to slow down. I often remember thinking that there must be an accident only to find that eventually the traffic lightens up and I never actually see the cause for the slow down.

Wednesday, April 13, 2011

Paper Reading #25: Tagsplanations: explaining recommendations using tags

Comments
Evin Schuchardt
Luke Roberts

Reference Information
Title: Tagsplanations: explaining recommendations using tags
Authors: Jesse Vig, Shilad Sen, John Riedl
Presentation Venue: IUI 2009: Proceedings of the 14th international conference on Intelligent user interfaces; February 8-11, 2009; Sanibel Island, Florida, USA

Summary
This paper discusses tagsplanations, explanations based on community tags that are made up of two parts: tag relevance (relationship of the tag to the recommended item) and tag preference (relationship of the user to the tag). They discuss an algorithm used to determine tag relevance and tag preference and present a study to see how users like tagsplanations.


Image taken from paper: Results sorted by tag relevance
To test the implementation of tagsplanations, the researchers use MovieLens, a movie recommendation site. They compute tag preference based on user behavior (by movie ratings in their MovieLens implementation). For example, to determine a user’s preference for the tag, violence, they would look at the user’s ratings of other films with the tag, violence, and calculate a weighted average for the preference value. They use the weighted average because some movies have been tagged with a certain identifier more than others and thus should be weighted more during the calculation. For tag relevance, they calculate the correlation between a user’s preference for a tag and their preference for the movie.

In their study the researchers sorted results in four different orders and then checked to see how well each interface helped users understand why an item was recommended and decide if they would like the recommended item. The researchers also checked to see if the recommended item matched their mood.

The results of the study showed that tag preference was more important than tag relevance but that tag relevance seemed to be the best way in which to organize the results. The interface that only used tag relevance (ordered by tag relevance and displaying tag relevance only) seemed to work the best for mood compatibility. They also found that subjective tags (tags expressing user opinions) performed better than factual tags (tags identifying facts about the movie such as concepts, people or places) overall.

Discussion
While I have found current recommendation systems to work just fine for me, I do think the researchers have presented some good ideas here that can make the entire system more informative and perhaps more relevant. In looking at their interface, I think they’ve done a good job of condensing a lot of information into a small amount of space.

Overall this was an interesting paper. They explained their motivations well and provided an in-depth user study. For future works the researchers mention making it so that users can inform the system when it is wrong and the system can inform users as to how different actions impact the results they receive. They also mention exploring other techniques to estimate tag preference.

Book Reading #51: Living With Complexity

Chapter 1
Summary
The first chapter serves as an introduction to complexity. Norman differentiates between things being complex (state of the world, the tasks we do and the tools we use) and the idea of things being complicated (the psychological state of a person attempting to understand, use or interact with something in the world). He discusses keys to coping with complexity, he brings up how some complicated systems occur as a result of poor design and he provides several examples of how we like complexity in certain instances (like when we already know a lot about the topic or where complexity truly seems appropriate). He also mentions that one way to measure complexity is by the amount of time it takes to learn and master a task (10,000 hours).

Discussion

This was a good first chapter on complexity. I think one of the most interesting things Norman points out is how musical instruments can be just as damaging to the hands as using a computer can be, but only computer companies tend to be sued for such injuries.

Chapter 2
Summary
In chapter 2 Norman brings up the idea of the conceptual model (the underlying belief structure held by a person about how something works), and how it helps to simplify the complexities of different systems. Norman discusses “featuritis,” adding more and more features, and explains how complexity is in the mind of the beholder. He also discuses how simpler looking does not always equate to simpler to use and how users tend to prefer an intermediate level of complexity.

Discussion
I liked the example he provided about how we as shoppers might take two items (like two types of toasters) and compare their features. Like he says, we’ve probably all done this. Usually I compare the features and the prices and ask myself if the additional features are worth the extra money. That’s what gets me thinking about whether I really need those extra features or not. Looks like I need to take that way of choosing an item one step further and really just think about what I need and not what has more knobs and buttons.

Paper Reading #24: Have a say over what you see: evaluating interactive compression techniques

Comments
Evin Schuchardt
Luke Roberts

Reference Information
Title: Have a say over what you see: evaluating interactive compression techniques
Authors: Simon Tucker, Steve Whittaker
Presentation Venue: IUI 2009: Proceedings of the 14th international conference on Intelligent user interfaces; February 8-11, 2009; Sanibel Island, Florida, USA

Summary
This paper discusses different Interactive Compression (IC) techniques to allow a user to remove information in documents they do not wish to see and making important information more visible.

The IC techniques they explore at different compression rates are:
1. Word Excision – remove unimportant words and replace them with periods
2. Utterance Excision – remove utterances (segments of de-emphasized information) and replace them with white space or an ellipsis for longer utterances
3. Highlighting Words – marking important words
4. Highlighting Utterances – marking important utterances
5. Keyword Context – use Utterance Excision but leave one word in place and grayed-out to let the user know the utterance is there
6. Fisheye View – split the screen into five views that display certain things such as important words, important utterances and segments of unmodified text


Taken from paper: Example of Word Excision
After doing an initial user study and finding that the two most successful IC techniques are Word Excision and Word Highlighting, they performed another more in-depth study on these two specific techniques.

Their results show that Word Excision and Word Highlighting allowed the users to extract the important information from documents more effectively even with the presence of an occasional error from the algorithm. Users were also found to prefer IC over unmodified text. There was no distinct preference for the level of compression used. It was different for each user. Word Excision allowed users to scan the document faster since words were omitted. For users who used Word Highlighting, they scanned the documents slower but did not lose any important information since nothing was omitted from the document.

Discussion
This paper was mostly about exploring current IC techniques. I can see it being referenced a lot in the future as other researchers delve deeper into this research topic. I think these different techniques could be very helpful. Personally, I would probably like the highlighting version the most, because it would allow me to still read information that is not highlighted if I want more information on a specific topic.

Recently I was assigned to read a paper about summarizing documents and now I had to read this one is about compressing the information provided. I think these areas of research go hand in hand and that the researchers could learn a lot by comparing their different findings and algorithms.

Areas of future study that researchers mention include improving their algorithms and IC techniques and exploring other ways to determine which parts of the document are important. They also want to test their system on other types of documents besides the type of documents used in these initial studies, documents from meetings.

Full Book Blog: Why We Make Mistakes

Summary
The Introduction introduces the ideas that are to be explored throughout the book. The author begins by describing mistakes and explains how much of the world around us is designed in such a way as to expect us to see things clearer and remember things better than we actually do. The author also defines “mistake” and outlines some of the topics to be discussed in the book: similar mistakes that happen, what a person can do to make fewer errors and understanding the role of context.


whywemakemistakes.com
Chapter 1 focuses on vision and how we don’t always see as much as we think we do. The author, Joseph T. Hallinan, gives several examples through movie mistakes and the door experiment, an experiment in which a person asking for directions switches places with another person passing through and helping carry a door. The author also explains the beer-in-the-refrigerator problem and how people have a quitting threshold when searching for something.

The second chapter continues the discussion of the first chapter, but doesn’t spend as much of the chapter on vision. The focus is on how the meanings we pick up on in what we see/hear are more important than the actual details of a scene. As in the first chapter, Hallinan gives many examples such as slip-of-the-tongue errors, recalling part of something (like a name) but not all of it and how we forget passwords and hiding places. At the end of the chapter Hallinan relates the discussion to crime and the process of identifying suspects.

Chapter 3 focuses on how we consider things. Hallinan begins with an example about people running for election and how voters make quick decisions based off of how competent a candidate looks. He gave another example of how people drinking wine rated the $90 wine as tasting the best when it was the same as the $10 wine – their brains even showed more evidence of being happy with the more expensive wine. Price is not the only thing that plays a part in this. The color of objects and a person’s memory of regret also play a role in people’s choices and consequently their mistakes. The author states that all of these examples are given to help readers better understand sources of mistakes.

Chapter 4 discusses how we tend to remember our actions with rose-colored glasses. When remembering things that we do and say, we tend to make ourselves sound better than we actually were/are. Hallinan presents several examples: students remembering their grades as being better than they were and remembering good grades more than the bad grades, the Watergate scandal and how John Dean remembered the events entirely differently than how they really went down (proven through tape recordings) and how gamblers remember their wins and how losses are near-wins in their eyes.

In chapter 5 Hallinan explores the idea of multitasking and how we don’t actually do it. He discusses how multitasking slows us down, can cause us to forget what we are doing and creates a need for downtime, the time it takes to refocus on a task. The author provides examples of a pilot, a bus driver and drivers in general who do other tasks (like texting, talking on the phone or fiddling with a GPS) while driving. He explores in depth the car and the many distractions being created to “aid” us distract us and how drivers need downtime when switching between tasks while driving.

In chapter 6 Hallinan discusses framing, how we look at something. He uses examples such as buying a certain wine depending on the music being played, the location of information regarding a company in a newspaper and the time that we make decisions (immediate or future) can affect our decisions. Hallinan also writes about multiple-unit pricing and the impact it has on shoppers. In seeing the tag, “4 for $2,” the number, 4, acts as an anchor that makes the shopper more likely to buy four of the item rather than just one or two. He relates this same idea to selling a house.

In chapter 7 Hallinan begins by showing us some examples of how we skim material and how this ability to skim comes with a trade-off: details are overlooked. He discusses how we skim when we read by only reading the first few letters of a work and assuming the rest, how a rookie piano player once noticed an error that had gone unnoticed for years and how a thirteen-year-old boy corrected NASA on their estimation regarding an asteroid.

Hallinan also discusses the importance of the context we are in when recognizing and remembering information. Hallinan demonstrates this with several examples: reading a description of doing the laundry before and after we know the context and a study in which people memorized words above and under water.

Chapter 8 focuses on a person’s desire to keep things organized even in memory. Hallinan begins with several examples to demonstrate this: people drawing the Seine River straighter than it actually is and stating where Reno, Nevada is in relation to San Diego, California. Hallinan further explains the idea of tidy memories through his discussion on how we like to organize information into a hierarchy. Hallinan also explains that who the person is will affect how he or she remembers things. In other words, people rationalize memories and change them, as shown in the study where people were asked to recall a folktale. When telling stories, people are also known to leave out details or make them up depending on the purpose of the story. These added or omitted details sometimes cause the person to remember the event differently.

Chapter 9 seemed to be a compare and contrast of men and women. Hallinan discusses the relationship between overconfidence and perceived risk. He also discusses how women seem to have less confidence than men in several areas. Examples he give includes getting lost while driving, selling back lottery tickets and fixing bugs in a spreadsheet. Hallinan relates some of these differences back to how boys tend to tinker more than girls, and boys tend to explore further out than girls.

Chapter 10 focuses on how people tend to believe they are above average. In other words, people tend to be overconfident. Hallinan sites many examples of this: golfing, paying for the gym on a yearly or monthly basis rather than day-to-day and credit card rates. Hallinan then discusses calibration, the difference between a person’s actual and perceived abilities. He discusses that calibration tends to be poor, but weather forecasts are well calibrated. The power of feedback is also discussed (feedback is poor when it comes to gym memberships). Hallinan also notes that as a task gets harder, people tend to be even more overconfident. The reason described for this is that people are overloaded with information and believe all the information actually makes them more likely to be right.

In chapter 11, Hallinan discusses how even professionals have difficulty in knowing how good or bad they are at something. He also discusses how experts become experts by practicing, starting at a young age and creating a library of specialized knowledge within the mind. Hallinan also presents information on a study of cognitive maps and mice, discusses how do-it-yourselfers hurt themselves with nail guns and how people prefer to do things without manuals in part because of how long the manuals can be. He then points out how people tend to do something in the first way they learned it – referred to as functional fixity – and backs up the idea with a task involving jars of water. Hallinan concludes by saying we should think outside the box though we usually do not when we already have learned a certain way to complete a task.

In chapter 12, Hallinan discusses constraints and affordances. He defines constraints as “simple mental aids that keep us on the right track by limiting our alternatives.” He defines affordances as “clues to how a thing can be used.” He discusses the technique by which different organizations name things and gives two examples: prescription drug names and identifiers that pilots use. He also discusses the importance of finding the root cause and knowing where to look (the culture of the place where we work) when searching for the source of an error. In discussing pilots and doctors, Hallinan also explains the effect that attitude has on errors.

In chapter 13 Hallinan begins with an example of two people who moved to L.A. and then ended up moving back to Wisconsin. He uses this example to show how people mispredict how they will feel about important life decisions in the future because they focus on relatively minor factors. He uses the same idea of misprediction to explain why gift cards are great for the company but not so great for the consumer.

In the Conclusion Hallinan gives some advice on how the readers can apply the ideas presented in this book to their lives. Some of the things he recommends we keep in mind are:

1. Think small
2. Calibration can be taught
3. Creating a written record helps a person fend off the rose-colored-glasses of hindsight bias
4. In some cases it will be important to look for and even expect failure
5. Don’t be set in your ways
6. Slow down
7. Be aware of the anecdote (he references NutriSystem here)
8. Get plenty of sleep
9. Be happy

Hallinan concludes by discussing how money does not eliminate mistakes and is not the currency of life.

Discussion
I really enjoyed this book and how the author presents so many examples to back up his claims. I especially liked the Conclusion, because Hallinan does a great job of summarizing his ideas without going overboard. It’s short and concise but effective. Some of the discussion in class was about how while this information is great, it is difficult to apply it to our lives to help prevent mistakes. While I still think this is true, I like how he gives some advice on how we can try to prevent the mistakes. I also think this book will be beneficial in helping us better understand mistakes we make, and as a Computer Scientist, I think this book will help us design better systems that prevent users from making certain expected mistakes.

Paper Reading #23: Improving meeting summarization by focusing on user needs: a task-oriented evaluation

Comments
Luke Roberts
Shena Hoffmann

Reference Information
Title: Improving meeting summarization by focusing on user needs: a task-oriented evaluation
Authors: Pei-Yun Hsueh, Johanna D. Moore
Presentation Venue: IUI 2009: Proceedings of the 14th international conference on Intelligent user interfaces; February 8-11, 2009; Sanibel Island, Florida, USA

Summary
This paper discusses a new approach to improving the summarization of meetings. The researchers discuss two types of summaries. The first provides a general summary of the meeting and is the type of summary produced by current systems (according to the paper, the general summary represents 30-40% of the meeting). The second provides a more decision-focused summary that is shorter than the general summary (according to the paper, the decision-focused summary represents 1-2% of the meeting).

The researchers performed a study in which they provided participants with four meetings through a Meeting Brower Interface. They asked the participants to summarize the decisions made in the meetings. Participants were randomly assigned one of four summary displays that were embedded into the browser interface that presented different information about the meeting. Some of the descriptions provided general summaries while others provided information generated by an algorithm described in another paper that pinpoints decision-related dialogue acts. In the study they focused on task effectiveness, report quality and user perceived success.

The researchers found that displaying decision-focused summaries were more effective and helped users get a better overview of the meeting. Even decision-focused summaries generated by the algorithm were more effective than general summaries created manually. However, it was found that decision-focused summaries written manually were still more effective than the ones generated by the algorithm.

Discussion
This sounds like it could be very useful for people who miss meetings or want to review meetings after they have occurred. As with most of the papers that I read in this class, I’m always left wishing I could know more about the actual algorithm they used, but the information provided in this paper was still interesting and proved that the algorithm and their interface can be useful for users.

As far as future studies, the researchers could focus on further improving the algorithm and they could run more experiments on meetings that are more and less structured to better identify the strengths and weaknesses in their current algorithm.
Image from paper of the Browser Interface