ALL ART BURNS

[I’m still building the web pages for the Official Documentation part of my Fall project that will include images, schematics, code, etc. I will post a link when that site is up. ]

For awhile now, I’ve been interested in physical and electronic security and the relationship they have with personal privacy. A large part of my day job for the past 8 years has involved privacy and computer security and I’ve been involved in some sort of security or hacking efforts since the 8-bit days. There are interesting areas of overlap where tradeoffs are made between security and privacy, and as I study design in a formal setting, I’m beginning to see the tradeoffs that will be made with design. While I’ve always been a proponent of security not getting in the way of usability, I didn’t worry too much about what “usability” actually meant and as a hacker, I was basically happy with with something that worked and was efficient.

In design school I’m learning design-thinking, which I’m discovering is similar to hacker-thinking (more on that in a future post). An interesting difference I’ve discovered is that as a hacker, I’m happy if it works. As a designer, I’m not happy until I’m no longer able to improve things. There’s a big leap between “it works well enough” and “it works and I can’t make it any better”, looking back at past hacking projects I can see now how much further I could have gone.

I’m also learning how to better communicate my ideas to people before I go off and act upon them. Hacking has always been about doing something, design is very much thinking about doing something and communicating the idea to people who do things. I want to do both. I want to show someone 20 variations on an idea using sketches, get useful feedback, then go off and do/make something to implement the idea.

For the Fall semester of mTID, I wanted to explore interaction design as it relates to learning about one’s physical and electronic surroundings. I didn’t want to go the AR route and wander around holding a mobile in front of my face or staring down into a tricorder; I wanted some sort of physical or tactile output device that feeds me information as an interrupt based on external triggers. Specifically, something that would let me know that I needed to be paying attention or that I should focus my attention in a specific direction but that was not always on at some idle setting or feeding me information that I have to parse as “negative/off”.

After looking at research in the field of haptic/tactile outputs, I also decided to take on the constraint of COTS technology, preferably open source hardware and software. I want other people to be able to replicate my work using free/cheap hardware and software instead of having to buy expensive mil-spec hardware or developing their own technology. There are teams of people with budgets doing some really interesting things, but I’m one person working in a self-funded studio.

Based loosely on what others have done, I ended up making a few waist-belts and arm-bands using Lilypad Vibe Boards driven by Arduino microcontrollers. The bands and belts are adjustable and can handle between 1 and 16 Vibe Boards. After getting all that up and running, I hooked it to a HM55B compass and did some basic navigation experiments and added a command-line interface to make it easy to manipulate individual Vibe Boards from outside the Arduino.

The thing that surprised me the most was the difference between what was reported in the literature and my personal experiences when it comes to using vibrators on a belt or band. Belts are a bit of a pain and have to be significantly readjusted and calibrated based on different individuals and the clothes they are wearing. For people who aren’t skinny, there are also issues with some of the vibe boards not making contact across the small of the back and for people who wear their pants at the wrong height it can often end up being a “torso band”. However, I was able to get some good practical experience with perceptual issues and concepts like the difficulty of supporting “just noticeable difference” between different wearers.

Due to the number of problems with the belt being used between different people, I started focusing on a band worn on the upper arm. My hope was that it would be easier to adjust and reconfigure, and that turned out to be the case. The arm-band also confirmed a problem that I’d first noticed during work developing the belt: vibrators are great for active feedback during a focused task or a high-priority alert but not so good for low-level alerts or the occasional nudge.

During debugging I noticed that I often startled myself by having a buzzer in my hand or resting on my skin go off, even when I was trying to figure out why it wasn’t powering up. Invariably, I’d drop the buzzer or flinch, even though I was expecting it to do something. (See also the classic “joy buzzer” prank.) My intuition is that we respond to surprise buzzing in a negative manner because of bees and other buzzing insects that can hurt us and that this is why almost every signal I’ve generated has seemed so negative if it happened unexpectedly.

However, having something seem like a surprise or creating a feeling of being startled is almost the exact opposite of what I’m trying to accomplish. My goal is to generate feedback at a less-critical level to alert a person of some information of interest, possibly including an abstract concept. I’m interested more in how to physically relate the sentence, “Hey, maybe you want to look in this direction for something suspicious that might also be hidden” and not the short imperative, “RUN! IT’S A LION!”

So now I’m looking at psych and HCI research again and considering combining vibrations with other tactile feedback. Perhaps a small solenoid that gets the wearer’s attention by “tapping them on the shoulder” then providing more information via the vibrators, or perhaps I can figure out a way to ramp up the vibrations slowly enough that someone isn’t startled.

As frustrating as this semester has been in terms of slow progress, setbacks, and dead-end tangents, I still feel like I’m learning a huge amount in a very short time. Once summer is here, I’m looking forward to some reflection time to sort out what I’ve really learned and how I can use it in the future.

Technorati Tags: arduino, design, haptic, ixd, lilypad, mTID