An accounting of how I spent the summer
Multiple times this year, I've had it banged into my head that the purpose of summer while one is in graduate school is to spend time raising one's research profile and working on projects that either require concentrated effort or is otherwise undo-able during the regular school year. In that spirit, I figured that an update to this neglected blog would be useful to let all three of my readers know what I've been up to.
This may be rather lengthy, so I'll be breaking things up by "project". A "project" in my mind is collection of tasks or goals focused on achieving a common end.
~
~
Graduate School
I've had a busy summer school-wise. My attention has been split between two main undertakings, my work with the Web Use Project and my work laying the foundations of my own research in human-computer interaction.
I've been working with Eszter Hargittai and the Web Use Project developing new tools and methods for surveying people in this wireless digital age. During the first half of the year, I developed a software application for making this methodology possible and toward the middle of spring, we began using the tool as a supplemental part of a larger data collection project. My work early in the summer was focused on aggregating the data collected and working with coders to tag the various responses. This went fairly well and we learned a lot. Toward the middle of the summer, Dr. Hargittai invited me to collaborate on a book chapter for a volume that she is editing that summarizes the work we did for an audience who may be attempting something similar in the future. We finished a draft of the chapter and we've received feedback from some readers. The next step is to incorporate this feedback and generate some further iterations of the chapter.
My own personal research has been focused on the topic machine learning in the context of ubiquitous computing. The best way that I can describe what I'm trying to accomplish is to create a system that acts much like a personal butler. This system serves as a gatekeeper between the user and the outside world, screening notifications and other incoming information in order to reduce the disruption that spontaneous interruptions cause.
The basic idea is this: someone sends you an e-mail. The e-mail application attempts to notify you with a "you've got mail" signal. The digital butler intercepts this notification and determines whether you are interruptible or not. If you are interruptible, the notification passes through normally. If you are not interruptible, the system either discards the notification or delays it until such a time you are interruptible.
There's a nice body of research that already addresses questions such as "when is the best time to interrupt users" (Shamsi & Bailey) and "how does the system learn when you are interruptible or not" (Fogarty & Hudson). At the moment, I am still aggregating the various research and trying to figure out how I can combine these various findings to create a usable system that can be measured in a lab performance-wise ("how well does the system effectively serve as a gatekeeper") and that can also be deployed out in the wild by regular users in regular contexts.
Aside from absorbing the relevant research, I've been spending quite a bit of time assembling my own toolbox of items that will assist me in building this system. The Growl plugin that I developed last month serves as a useful canvas for experimenting with different kinds of notification styles and modalities. I've developed an application called "Do Not Disturb" that collects various readings from various sensors that will be fed into a machine learner that will sense users' context. I began dabbling in the home automation space with Shion to create an application that is not notification-oriented, but can still uses the context-sensing machine learner in a novel and useful manner. ("Chris is now working in front of the computer - open the blinds and let in energizing natural sunlight.")
In addition to all of this research reading and original coding, I've been reading a couple of HCI books a month in order to learn more about the field and to not fall into an overly specialized hole of my own creation.
I've had a busy summer school-wise. My attention has been split between two main undertakings, my work with the Web Use Project and my work laying the foundations of my own research in human-computer interaction.
I've been working with Eszter Hargittai and the Web Use Project developing new tools and methods for surveying people in this wireless digital age. During the first half of the year, I developed a software application for making this methodology possible and toward the middle of spring, we began using the tool as a supplemental part of a larger data collection project. My work early in the summer was focused on aggregating the data collected and working with coders to tag the various responses. This went fairly well and we learned a lot. Toward the middle of the summer, Dr. Hargittai invited me to collaborate on a book chapter for a volume that she is editing that summarizes the work we did for an audience who may be attempting something similar in the future. We finished a draft of the chapter and we've received feedback from some readers. The next step is to incorporate this feedback and generate some further iterations of the chapter.
My own personal research has been focused on the topic machine learning in the context of ubiquitous computing. The best way that I can describe what I'm trying to accomplish is to create a system that acts much like a personal butler. This system serves as a gatekeeper between the user and the outside world, screening notifications and other incoming information in order to reduce the disruption that spontaneous interruptions cause.
The basic idea is this: someone sends you an e-mail. The e-mail application attempts to notify you with a "you've got mail" signal. The digital butler intercepts this notification and determines whether you are interruptible or not. If you are interruptible, the notification passes through normally. If you are not interruptible, the system either discards the notification or delays it until such a time you are interruptible.
There's a nice body of research that already addresses questions such as "when is the best time to interrupt users" (Shamsi & Bailey) and "how does the system learn when you are interruptible or not" (Fogarty & Hudson). At the moment, I am still aggregating the various research and trying to figure out how I can combine these various findings to create a usable system that can be measured in a lab performance-wise ("how well does the system effectively serve as a gatekeeper") and that can also be deployed out in the wild by regular users in regular contexts.
Aside from absorbing the relevant research, I've been spending quite a bit of time assembling my own toolbox of items that will assist me in building this system. The Growl plugin that I developed last month serves as a useful canvas for experimenting with different kinds of notification styles and modalities. I've developed an application called "Do Not Disturb" that collects various readings from various sensors that will be fed into a machine learner that will sense users' context. I began dabbling in the home automation space with Shion to create an application that is not notification-oriented, but can still uses the context-sensing machine learner in a novel and useful manner. ("Chris is now working in front of the computer - open the blinds and let in energizing natural sunlight.")
In addition to all of this research reading and original coding, I've been reading a couple of HCI books a month in order to learn more about the field and to not fall into an overly specialized hole of my own creation.
~
Recreation
In terms of recreation, I've been staying up to date on the latest summer movies and I've been making regular trips to the comic store. I've made significant progress in filling the holes in my X-Men collections and I've picked up a few new titles to read regularly. (Peter David's "X-Factor" may be my favorite summer find.)
Aside from World of Warcraft, I haven't been gaming all that much. My most played non-WoW game is probably Desktop Tower Defense. I did pick up "Final Fantasy XII", but it has yet to really hook me and suck me into the game.
In World of Warcraft, I've been mostly farming all summer. Earlier this year, I finished most of the quests in the expansion and became bored with the game. With a recent patch, Blizzard introduced a new quest line for my class (Druid). The downside was that it required that players have obtained their epic flying mount, which costs 5200 gold. This is an enormous sum and I've been farming herbs and playing the Auction House in order raise the funds. I haven't reached my goal yet, but I should within the next couple of days.
In terms of reading, I've gone through a variety of books. However, my most significant find this summer was Robert E. Howard and his Conan stories. I have been working through a thousand page Conan compilation and taking side trips into the comics. I've been enjoying what I've been reading and look forward to finishing the compilation one of these years. In addition to Conan, I've been listening to Robert Ludlum's "The Sigma Protocol" and earlier this month, I finished the latest installment of Katherine Kerr's Deverry series.
Recreation
In terms of recreation, I've been staying up to date on the latest summer movies and I've been making regular trips to the comic store. I've made significant progress in filling the holes in my X-Men collections and I've picked up a few new titles to read regularly. (Peter David's "X-Factor" may be my favorite summer find.)
Aside from World of Warcraft, I haven't been gaming all that much. My most played non-WoW game is probably Desktop Tower Defense. I did pick up "Final Fantasy XII", but it has yet to really hook me and suck me into the game.
In World of Warcraft, I've been mostly farming all summer. Earlier this year, I finished most of the quests in the expansion and became bored with the game. With a recent patch, Blizzard introduced a new quest line for my class (Druid). The downside was that it required that players have obtained their epic flying mount, which costs 5200 gold. This is an enormous sum and I've been farming herbs and playing the Auction House in order raise the funds. I haven't reached my goal yet, but I should within the next couple of days.
In terms of reading, I've gone through a variety of books. However, my most significant find this summer was Robert E. Howard and his Conan stories. I have been working through a thousand page Conan compilation and taking side trips into the comics. I've been enjoying what I've been reading and look forward to finishing the compilation one of these years. In addition to Conan, I've been listening to Robert Ludlum's "The Sigma Protocol" and earlier this month, I finished the latest installment of Katherine Kerr's Deverry series.
~
Books
Speaking of "books", this summer has seen quite a bit of work on my Books application. Last week, we released the first major update to the 3.x codebase in version 3.1.0. This version had been under development for a couple of months and introduced some improvements and new features to the application. The largest new feature is the new Gallery View that was my first foray into raw NSView programming on the Mac. I learned quite a bit doing this and I've been applying these lessons in other applications, such as Shion.
Exercise & Personal Care
Speaking of "books", this summer has seen quite a bit of work on my Books application. Last week, we released the first major update to the 3.x codebase in version 3.1.0. This version had been under development for a couple of months and introduced some improvements and new features to the application. The largest new feature is the new Gallery View that was my first foray into raw NSView programming on the Mac. I learned quite a bit doing this and I've been applying these lessons in other applications, such as Shion.
~
Exercise & Personal Care
While I've been a bum with respect to working out at the gym during the month of August, I was exercising regularly earlier during the summer. I've fallen off of the horse, but I've given myself a few more days before beginning again in earnest. I've been enjoying the morning exercises and look forward to resuming them once September rolls around.
In other aspects of personal care, I've adopted Merlin Mann's Inbox Zero system for managing e-mail. I've been living in iGTD to stay on top of things and I've gone through a few cleanings around the house to shrink the clutter for my own mental well-being. (It's been a good year for donation places.)
During the middle of August, Holly and I traveled out west to see my brother's baby girl, visit my family, and to attend Lucille's wedding. The baby was adorable, if aloof. The family was doing fine. And Lucy had a very nice wedding, even when accounting for a questionable speech by the best man. :-)
In other aspects of personal care, I've adopted Merlin Mann's Inbox Zero system for managing e-mail. I've been living in iGTD to stay on top of things and I've gone through a few cleanings around the house to shrink the clutter for my own mental well-being. (It's been a good year for donation places.)
During the middle of August, Holly and I traveled out west to see my brother's baby girl, visit my family, and to attend Lucille's wedding. The baby was adorable, if aloof. The family was doing fine. And Lucy had a very nice wedding, even when accounting for a questionable speech by the best man. :-)
~
As the summer months passed, I couldn't help myself from thinking, "I can't believe it's already June/July/August." However, now I'm looking forward to the fall and the resumption of classes and a more regular schedule. Looking back, I'm happy with what I've accomplished over the past few months, and I don't have any regrets about how I've spent my time. From my perspective, I've worked hard and played hard, and in the end that's all that I can ask for.
