Category: RestOfYou

Last week, we were to add a layer of polish to one of the three assignments (log self, log environment, log computer). I opted to add something to the third piece by analyzing more words and visualizing that. In addition to counting my logged keys for a week, I used eclipse to count words from my ITP blog entries. The results were a little dry, but surprising in quantity and range. (I came across some interesting sites including wordnet). I also uploaded my data sets to a fellow ITPer’s thesis project site called openvisuals. While the visualization options did not fit my data as nicely as I would have liked, I thought this stringed pie chart worked ok (it’s interactive on the site itself).

What I was thinking about more so though was a final project — how I could really get into the core of the course and produce something both engaging and tangible. The ideas I brought to the class seemed like a fog in my mind, but the discussion that resulted helped me to choose a direction to take. In short, I want to be able to extract my thoughts, steadily, with minimal effort, and minimal immediate intention. Why? I’m not sure exactly, but I’m thinking about quality of life and self awareness. The assignments so far have given us several different perspectives and data that I haven’t really seen before, but the most subtle things and perhaps most interesting things are the ones that aren’t explicit, aren’t communicated, and perhaps require the most digging to reveal. In my case, my dreams and internal monologue / haze would be great to see and experience outside of my head — being able to measure from excitement, to confusion, to depression, to critical analysis, etc. Not only would it be enlightening in and of itself, but it would have numerous practical applications. It could maybe take journaling or therapy to a new, private, empowering level.

The plan was to record my thoughts through voice (since writing or typing requires more labor and possibly more intention). Then I would try to process the voice with two kinds of software: 1) transcription (content) and 2) stress analysis (abstract). The plan seemed challenging but great.

Reporting back: I checked out an m-audio kit from the equipment room, recorded several streams of thought, and saved them as mp3 files. Meanwhile, I read-up on transcription software and found MacSpeech for OSX (equivalent to Naturally Speaking). I installed this and began using it to dictate. My laptop’s internal mic is messed up, so I borrowed a snowball mic from the equipment room to train the software to recognize my voice. It works well, but getting the software to transcribe from the mp3 files is the technical problem I’m experiencing now. I’m using soundflower and line-in to redirect the audio in and out, but MacSpeech won’t seem to allow me to apply the settings from one profile to another… I haven’t touched any stress analysis software yet, but seems this will be the more complicated of the two even. Another issue is that a large portion of the thoughts I’d like to see would not be able to be spoken, because of my environment, whether it be another person, a group, a shower, train noise, etc. While the potential of this project is highly attractive, I’m not sure it’s going to be feasible.

The first two segments of the semester focused on “logging yourself” and “logging your environment.” The most recent segment focused on “logging your computer.” Some of my peers tracked and visualized their mouse movements, some were sniffing packets, and others logged their keystrokes. I fell into the latter group, using a program called logkext to keep a record of my typing. The advantages of the program are that it’s lightweight, free, runs in the background, and encrypts the log file. Still, there is a bit of risk in logging key strokes because of sensitive information including login details and private conversations. Regardless, I installed the client and ran it for about a week. The terminal interface was simple and refreshing to use.

Once I gathered the data, I found the first step in analyzation (word countring) to be tricky, but fortunately our helpful instructors at ITP provide code to move that process along (see Daniel Shiffman – concordance and Dan O’Sullivan – language). Even with the code, getting the raw data into an organized list of counted words wasn’t obvious for a eclipse coding novice, but I got to the stage I wanted. **Note: in Eclipse, there’s an option when you select “run as” called “run configurations,” which allows a user to specify a file path in the “(x)=arguments” tab. You can see the top 500 words here. To get that nice round number, I manually removed noise and cropped off anything typed less than twice. The next step with this project would be to compare this data to other data …

After saving the photos from my first attempt at snapping about a day’s worth of photos, I used the Open Image Sequence option in Quicktime to make a time-lapse video. I tried various frame rates, and found 10 fps to be a fair balance for me in this case. I added some audio with iMovie as well: Millenium from Outkasts’s ATLiens, because I was listening to this album for a good portion of the time while the camera was capturing images. Check it out below.
snap-a-day from ben yee on Vimeo.

My intentions moving forward would be:

• repeat the process with higher resolution photos and shorter intervals, perhaps 1024×768 and an image every 5 seconds.
• capture a day that includes more interesting content (i.e. beyond monitoring my laptop screen)
• include an object that remains consistent for reference (see youtube example)
• experiment with different angles
• analyze scenes more carefully, noting the relationship between activity duration and level of happiness, interest, excitement…
• insert music samples or sound effects to create a multi-layered story of sorts
• continue ideation

As of my last entry, the plan for the enviroment-logging-assignment has changed a bit. I’ve learned that Blackberry devices are relatively locked down, and therefore difficult to use for creative projects. Rather than trying to hack the Blackberry, a beautiful solution occurred to me — why not borrow a Symbian device and try loading an existing application readily available online. Smile. With some generosity from ITP, I now have an N95 in hand, with 8gb of memory already available (another 8 gb chip and ipod shuffle are on the way via dhl still). After some reading and tinkering, I’ve concluded that N series devices are pretty dope…free connection through wifi, Bluetooth giving easy access to the images. So, I installed an application called Snapper written by Markus Mertama. Super simple and free. I started the capture around 2 pm today, and I’m shooting 640×480, every 20 seconds, just above waist level. I’ve been sitting in front of my machine most of the day, working with this for an hour or two to get it going, but then moreso coding a little Dating Game Show project for LiveWeb. I’ll keep shooting pics tonight and then again tomorrow morning. So far, I have around 1000. This should be cool.

Part of the assignment for the week is to read a study about American youth and flow/happiness here. I’ve read it through once and am about to give it a second pass in order to get a stronger understanding and opinion. At this point though, I recall the study to be engaging, but the method and the read itself to be semi-tedious…

2nd go: While some of the study seems to fall into the realm of common sense: “teenagers are happier when they do certain things (e.g. in leisure) than when they do others (e.g. study)”, it also introduces many interesting points of discussion. I found the excerpts below to be noteworthy.

• Hope, optimism and the ability to experience flow can be learned and thus moderate one’s level of happiness (Csikszentmihalyi, 1997; Seligman, 2002).
• Experience Sampling Method (ESM) … relies on subjects’ responses to an electronic pager that signals at random times during the waking hours of the day, yielding up to fifty measures of happiness at specific moments during an average week.
• To measure whether a person was more likely to be in a Flow condition we calculated the percent of time spent in situations that were above the mean level of challenge and the mean level of skill at the same time. When a person was above the mean of skills but below mean challenge, the condition was considered conducive to Relaxation. High challenges and low skills were counted as Anxiety, and low challenges with low skills as Apathy.
• Discussion from pages 196-198.

The other part of the assignment is to plan an environmental logger. The Logomatic and Logomatic v2 from Sparkfun seemed like the most simple solutions for the task but are currently out of stock. Fortunately, this misfortune caused me to take a closer look at several of the links from the class blog, including WayMarkr, Sensecam, UCLA, Biomapping, Off-the-shelf. So now I’m in a predicament: uncertain hardware plan and possibilities aiding my indecisive nature. I really just want a Sensecam… hopefully tomorrow’s class will give insite and I’ll still be able to get supplies soon enough.

A sidenote while looking at other sensors on Sparkfun and thinking about trusting my body as a sensor of sorts: So yes, we’re tracking our environment beyond our body, beyond what we see, smell, hear, feel, taste, (and remember), so as a humans, we are equipped with some pretty awesome physical “tools.” A thought entered my mind though, which is nothing new (ie the existence of a bio-tech industry), but I guess new to me — What if we could upgrade a component of our body in the same way we upgrade our devices? Corrective lenses, earpieces, etc aside, I’d like some more memory and a nicer internal filing system. Better battery life wouldn’t hurt either.

* I’ve decided to go with a Blackberry Pearl for my sensing device, which will take a days worth of time-lapse photography. I’m thinking of positioning it at waist level to give an idea of what a child would see. Power issue – check. Storage issue – to buy microSD card. Functionality – to research today.