4
Summary
This chapter covers reflection; that is, external aids to thought. It is fairly detailed, with an emphasis out contradicting intuitive wisdom, or what seems like such today.
Discussion
The strongest chapter of this book so far, making a bunch of good points. I had difficulty with tic tac toe example (even though I knew that was what it was, I couldn't remember which value went to which square). I did, however, find the paragraph more helpful than the visual on page 61.
5
Summary
This chapter covers much the same material as the previous, only from the other end. That is, he discusses how artifacts can be made optimal for humans, rather than the way humans interact with artifacts.
Discussion
This was another strong chapter, although I've heard all the affordance stuff before. I wonder if the televisions of 1993 worked differently than today; my television has a perceptual difference between "off" black and "no picture" black.
Tuesday, April 5, 2011
Paper Reading #20, "Lowering the Barriers to Website Testing with CoTester"
http://csce436-hoffmann.blogspot.com/2011/04/paper-reading-19-vocabulary-navigation.html
http://aaronkirkes-chi.blogspot.com/2011/04/paper-reading-19-personalized-news.html
Lowering the Barriers to Website Testing with CoTester
Jalal Mahmud and Tessa Lau, IBM
IUI 2010, 7-10 Feb '10, Hong Kong
Summary
CoTester is, in the author's words, "a lightweight web testing tool which can help testers easily create and maintain test script". The intention, if I understand correctly, is to have a tool that can be used for easy, automated testing of website functions. The authors extended an existing, easy to learn scripting language (CoScripter) for the project, with the goal of creating a script testing tool that did not require knowledge of Java/Visual Basic to utilize.
I'm afraid that the implementation, which they went into a quite detailed explanation of, was a bit beyond my level and I do not feel I can relate it properly. Interested readers should refer to the paper.
The results were quite promising. The script did an excellent job of identifying problems, exceeding the comparison algorithm's success rate by 14% (91% to 77%), and using cosine similarity scores over straight up equality checking to determine which class instructions belong in was also quite successful.
Discussion
The item that jumped out at me here as questionable - and I may be a little off-base here - was that I'm not sure I want my debugging script to be written by someone without at least a minimal programming background. I mean, Java and Visual Basic aren't exactly the most difficult languages in the world. Still I'm sure there's some application for this and they seem to be getting good results.
A good automated testing system for anything development related could of course be of great benefit to any programmer, just ask the Extreme Programming guys.
I agree with the authors that a much-expanded user study would be a good next move, but I'm definitely not qualified to tell them what to do next in the technical sense.
http://aaronkirkes-chi.blogspot.com/2011/04/paper-reading-19-personalized-news.html
Lowering the Barriers to Website Testing with CoTester
Jalal Mahmud and Tessa Lau, IBM
IUI 2010, 7-10 Feb '10, Hong Kong
Summary
CoTester is, in the author's words, "a lightweight web testing tool which can help testers easily create and maintain test script". The intention, if I understand correctly, is to have a tool that can be used for easy, automated testing of website functions. The authors extended an existing, easy to learn scripting language (CoScripter) for the project, with the goal of creating a script testing tool that did not require knowledge of Java/Visual Basic to utilize.
I'm afraid that the implementation, which they went into a quite detailed explanation of, was a bit beyond my level and I do not feel I can relate it properly. Interested readers should refer to the paper.
The results were quite promising. The script did an excellent job of identifying problems, exceeding the comparison algorithm's success rate by 14% (91% to 77%), and using cosine similarity scores over straight up equality checking to determine which class instructions belong in was also quite successful.
Discussion
The item that jumped out at me here as questionable - and I may be a little off-base here - was that I'm not sure I want my debugging script to be written by someone without at least a minimal programming background. I mean, Java and Visual Basic aren't exactly the most difficult languages in the world. Still I'm sure there's some application for this and they seem to be getting good results.
A good automated testing system for anything development related could of course be of great benefit to any programmer, just ask the Extreme Programming guys.
I agree with the authors that a much-expanded user study would be a good next move, but I'm definitely not qualified to tell them what to do next in the technical sense.
The most accessible illustration from the paper. Pictures were not a strong suit.
Sunday, April 3, 2011
Ethnography Results, Week 8
I went back to the group that meets Saturday night at McDonalds. It was post-convention tabletop games night, so there wasn't any D&D. Nothing else Earth-shattering to report.
Microblog #44, "Why we Make Mistakes, Chapters 4&5"
4
Summary
This chapter is about how hindsight isn't as clear as we think it is. It shows several statistical studies of people misremembering to put themselves in a better light.
Discussion
This is...interesting. I'm familiar with the sports gambler phenomena (fantasy football, etc) but I find it tends to be far less prevalent among statistically minded people (who generally, are also the best at predictions). I wonder if this is generally true.
Summary
This chapter is about how hindsight isn't as clear as we think it is. It shows several statistical studies of people misremembering to put themselves in a better light.
Discussion
This is...interesting. I'm familiar with the sports gambler phenomena (fantasy football, etc) but I find it tends to be far less prevalent among statistically minded people (who generally, are also the best at predictions). I wonder if this is generally true.
5
Summary
This chapter talks about how multitasking isn't really, with an emphasis on in-car distractions.
Discussion
Visual distractions are the worst problem for drivers; at some point we'll have voice systems good enough that eyes can be kept on the road while tasks are being carried out. I think the calls for more regulation are a little overblown; at some point, writing unenforcible laws just make you look silly (leaving aside entirely liberty-vs-security/safety concerns).
Microblog #43, "Things that Make us Smart, Chapter 1&2"
1
Summary
This chapter opens Dr. Norman's book Things that Make Us Smart. The two principal themes are over-entertainment at the expense of education (for example, TV news vs. newspapers, more flash but generally less content) and machines working more for machines than for people (user unfriendly interfaces failings being entirely brushed off on people).
Discussion
If it sounds like you've heard this before, it's because you have. This chapter has a little bit more of a rambling quality than Norman's previous work, we'll see if that continues.
2
Summary
This chapter expands on the first, then describes the three kinds of learning (accretion, tuning, restructuring) and explains the phenomena of optimal flow.
Discussion
I must admit that I was not in "optimal flow" as I was reading this. I did, however, note that he's still trying to reverse engineer things from the playstation (pg. 22).
Summary
This chapter opens Dr. Norman's book Things that Make Us Smart. The two principal themes are over-entertainment at the expense of education (for example, TV news vs. newspapers, more flash but generally less content) and machines working more for machines than for people (user unfriendly interfaces failings being entirely brushed off on people).
Discussion
If it sounds like you've heard this before, it's because you have. This chapter has a little bit more of a rambling quality than Norman's previous work, we'll see if that continues.
2
Summary
This chapter expands on the first, then describes the three kinds of learning (accretion, tuning, restructuring) and explains the phenomena of optimal flow.
Discussion
I must admit that I was not in "optimal flow" as I was reading this. I did, however, note that he's still trying to reverse engineer things from the playstation (pg. 22).
Full Blog, "Coming of Age in Samoa"
Coming of Age in Samoa
Margaret Mead
William Morrow & Company, USA, 1928
Summary
This book is an ethnographic study of youth, especially girls, conducted in 1920s Samoa by Dr. Margaret Mead in the hopes of finding a population of that age range that wasn't caught up in the "noise" of Western society. She spends fourteen chapters discussing the results of her study, then follows it up with a statistical appendix.
Samoa is described as a very laid back culture, with the worst of the previous primitive culture having been eliminated by Western contact without having having yet acquired the hectic state of the modern West. Dr. Mead's report is summarized in chapters 13 and 14, with 13 being more a real summary and 14 being a soapbox speech on its applicability to the US of the 1920s.
Discussion
The significance of this book hardly needs to be expounded upon, as it is considered the classic of the genre. The two things I really didn't like about were 1. there was far more information in some sections than I wanted, as I'm sure can be recalled from my commentary on the appropriate chapters, and 2. I strongly disagree with the final chapter. Dr. Mead's ideas on child rearing are anathema to me. For future work, at this point I could just pull up some materials on modern Samoa.
Margaret Mead
William Morrow & Company, USA, 1928
Summary
This book is an ethnographic study of youth, especially girls, conducted in 1920s Samoa by Dr. Margaret Mead in the hopes of finding a population of that age range that wasn't caught up in the "noise" of Western society. She spends fourteen chapters discussing the results of her study, then follows it up with a statistical appendix.
Samoa is described as a very laid back culture, with the worst of the previous primitive culture having been eliminated by Western contact without having having yet acquired the hectic state of the modern West. Dr. Mead's report is summarized in chapters 13 and 14, with 13 being more a real summary and 14 being a soapbox speech on its applicability to the US of the 1920s.
Discussion
The significance of this book hardly needs to be expounded upon, as it is considered the classic of the genre. The two things I really didn't like about were 1. there was far more information in some sections than I wanted, as I'm sure can be recalled from my commentary on the appropriate chapters, and 2. I strongly disagree with the final chapter. Dr. Mead's ideas on child rearing are anathema to me. For future work, at this point I could just pull up some materials on modern Samoa.
Saturday, April 2, 2011
Paper Reading #19, "WildThumb"
http://chiblog.sjmorrow.com/2011/03/paper-reading-19-tell-me-more-not-just.html
http://csce436spring2011.blogspot.com/2011/03/paper-reading-19-local-danger-warnings.html
WildThumb: A Web Browser Supporting Efficient Task Management on Wide Displays
Shenwei Liu, Cornell University
Keishi Tajima, Kyoto University
IUI’10, February 7–10, 2010, Hong Kong, China
Summary
This paper describes a new system designed by the authors to allowing easier tab switching than the currently available systems for web browsers on wide screen monitors. It uses the extra space to display "augmented thumbnails" instead of traditional tabs, making the pages more visible and easier to click.
The thumbnails themselves consist of an image of the top of the page, with the site logo overlaying the upper left and the most prominent image on the page overlaying the lower right. The basic idea is clearly illustrated in the image below, taken from the paper.
A 9-user study conducted with experienced web browser users led to the conclusion, both from timing operations and from questionnaires issued to the subjects - that the system provided for a minor increase in switching speed.
Discussion
This idea was very interesting (which I think will be illustrated by the length of this discussion section!) if, in my view, somewhat flawed. The idea of making improvements to the tab system is of course broadly applicable and would be quite useful to anyone.
The concerns I have are as follows. First, look at the above screenshot. While the contents of the unopened tabs are more clear, the trade off is that they are also quite distracting, drawing the eye away from the primary focus. Second, the excess space being utilized here is going to vanish as more and more websites allowing widescreen browsing, meaning that you will be paying an increased price in terms of readability to get these augmented sidebar thumbnails. I am also somewhat concerned about the auto-page grouping algorithm, as I prefer to maintain control over the positioning of my tabs myself and find that the widely available click and drag functionality is quite adequate for this. This concern could be allayed by simply allow that algorithm to be toggled off. I must also note that I have caught the first grammatical error I can recall in one of these papers: in the 2nd sentence of the introduction "are" should be "is".
On a positive note, the augmented thumbnails do live to their billing, and could improve many different pages/functions that use thumbnails. The Chrome homepage, for instance.
In the future, two functions I want to see are the ability to view two different tabs from the same browser at one time (presumably, each taking up half the screen by default), and the ability to have a function otherwise similar to "favoriting" save and open multiple tabs at once.
http://csce436spring2011.blogspot.com/2011/03/paper-reading-19-local-danger-warnings.html
WildThumb: A Web Browser Supporting Efficient Task Management on Wide Displays
Shenwei Liu, Cornell University
Keishi Tajima, Kyoto University
IUI’10, February 7–10, 2010, Hong Kong, China
Summary
This paper describes a new system designed by the authors to allowing easier tab switching than the currently available systems for web browsers on wide screen monitors. It uses the extra space to display "augmented thumbnails" instead of traditional tabs, making the pages more visible and easier to click.
The thumbnails themselves consist of an image of the top of the page, with the site logo overlaying the upper left and the most prominent image on the page overlaying the lower right. The basic idea is clearly illustrated in the image below, taken from the paper.
A 9-user study conducted with experienced web browser users led to the conclusion, both from timing operations and from questionnaires issued to the subjects - that the system provided for a minor increase in switching speed.
Discussion
This idea was very interesting (which I think will be illustrated by the length of this discussion section!) if, in my view, somewhat flawed. The idea of making improvements to the tab system is of course broadly applicable and would be quite useful to anyone.
The concerns I have are as follows. First, look at the above screenshot. While the contents of the unopened tabs are more clear, the trade off is that they are also quite distracting, drawing the eye away from the primary focus. Second, the excess space being utilized here is going to vanish as more and more websites allowing widescreen browsing, meaning that you will be paying an increased price in terms of readability to get these augmented sidebar thumbnails. I am also somewhat concerned about the auto-page grouping algorithm, as I prefer to maintain control over the positioning of my tabs myself and find that the widely available click and drag functionality is quite adequate for this. This concern could be allayed by simply allow that algorithm to be toggled off. I must also note that I have caught the first grammatical error I can recall in one of these papers: in the 2nd sentence of the introduction "are" should be "is".
On a positive note, the augmented thumbnails do live to their billing, and could improve many different pages/functions that use thumbnails. The Chrome homepage, for instance.
In the future, two functions I want to see are the ability to view two different tabs from the same browser at one time (presumably, each taking up half the screen by default), and the ability to have a function otherwise similar to "favoriting" save and open multiple tabs at once.
Subscribe to:
Posts (Atom)