As I’ve noted in earlier documents, my thesis plan is to create a series of devices which allow people to explore the urban spaces around them through the use of ambient monitoring devices. These devices will be built, tested and disributed to users in stages — partly so I can start with lower-tech ideas and get the first prototypes out faster, and also so I can make adjustments to later ideas based on feedback on the first rounds. I’ve outlined the design treatement for the devices in the following pages, starting small and building up.
Personal Monitor
Look & Feel
The first object I plan to develop is a personal sound monitor for city-dwellers to use in their daily routines. This will be a simple device, technically speaking, with a microphone, LED, and microcontroller to read one and control the other. When nothing is happening, it will pulse a very faint white, a sort of “everything’s okay” indicator. It will listen for levels of sound, and when it detects something over the acceptable level (as determined by the user, assumedly starting with an “unhealthy” volume), it will go bright red, and fade back to white over a period of time determined by the user (the default will be half an hour, and can be lengthened to an hour or two and shortened to less than a minute).

The idea here is that it’ll be small enough to stick in a pocket, bright enough to always be visible, and feel comfortable and calming. To me, this means housing the device in a smooth, translucent plastic case, so that the light can be seen clearly but the electronics cannot — this should provide the necessary information without cluttering. The monitor, on the inside, would have two knobs: one two control the decay time and the other to control the mic sensitivity, as well as all the technical guts: an ATMega168 (or ATTiny) bootloaded with Arduino software, a microphone element with preamp circuit, battery, and LED (along with various other capacitors and resistors to smooth out current and prevent resets). And, as noted, all this will have to be put in a pleasant-feeling case, so I’ll try both casting my own cases and using off-the-shelf plastics to see what people find more calming and personal.

Future development
Once it’s all working and tested, the first goal is to shrink the size as much as possible. If I can get the whole circuit to be the size of a tealight candle (or smaller), I think I’ll have won — and I don’t think that’s out of the question. I’m also toying with the idea of putting in a small vibrating motor, just to call additional attention to damaging sound events through ambient alerts. It’s more mentally demanding than my previous idea, and requires the user to pay attention, but perhaps some might prefer that. Either way, if a vibration element is added, I’d include an on-off switch for it.
Eventually, I’ll build similar units to watch for other elements and display other colors — at the moment, I’m thinking green for gases and blue for vibration (and I’m toying with purple for completely made-up “paranormal activity” detection, though we’ll see on that). Once the sound part is done and tested, this should be relatively easy to set up — just add the proper sensor, change the light color, and go. I’m considering whether it’s better to have one light do all the changing, or have separate lights — have to see what’s most informative / least confusing for users.
If I have time to visit this option, I’d like to explore logging the data of the day, though I’m leaving that part aside for now. I think that if I include an internal clock/timestamp (or tie it to a mobile phone which collects the data) I can get enough interesting data to make it informative for the user, but I don’t want to drown in that part just yet — and others are certainly taking that approach as well.
Concerns / issues to address
What I worry is that the object may be lost on people — either they’ll forget they have one, or they will simply not find the information useful. Of course, cost is a concern here as well, and will have to be addressed. And finally, this is an honest concern for me: I’m on one hand picturing the personal monitor as a wearable, on the other viewing it as a leave-behind, and picturing it being viewable from relatively far away while not bothering users at close range. This may involve dimming switches, but more likely will involve making two separate classes of devices, which leads me to:
Public Markers / Initial Wearable
Look & Feel
I’m looking into making much smaller, minimally-techy versions of the first monitor as part of the leave-behind idea. As I’ve noted in earlier documents, I’m hoping to make small monitors that simply hear sounds and react with light (and later, other sensors/senses as well). The leave-behinds will be purposefully simple initially, partially because I don’t want to drown in technology while building prototypes, but also because I’d like this to be the easiest one for people to make on their own. These will have few parts and less costly elements, partly to increase the battery life, partly to make them easier to reproduce, and partly to make them cheap enough that if a few go missing, it won’t kill me.

I envision people carrying a few with them, then “tagging” public spaces to display sound change in a form of “informative graffiti” (I still hate that term, but it works). Ideally, once a few are up, passers-by will be able to get a feel for how loud an area has been and get some idea of what may have happened there and how long ago, just by looking up and seeing the various shades of color.

Further Development
I’ll be tinkering, as outlined above, with size and tech first, but eventually also working in monitoring and wireless abilities to eventually make a networked wearable. As I see it, these could be little monitors or completely embedded in clothes/personal items — they just need to be visible from a short distance. These will do the same monitoring on their own, but will also have a wireless radio to listen for other people’s wearables. So, ideally, if you walk out of your apartment in the morning, yours will be calm and white, but when you exit the subway it will likely be bright red, displaying to other people that you’ve just been somewhere loud — and ideally, they’ll take action based on that information. Additionally, these devices will talk —when it detects another of its kind that has a high affectation, it turns a little reddish to show empathy, and the damage that sound causes to social situations (and provides data for the reverse situation). This, obviously, makes even more sense with the gas sensor idea.
Concerns / issues to address
The biggest issue I’ll have to address is for the ideas outlined above, which functionality is best suited to which. It may turn out that the monitor makes the better wearable idea, while the public markers should remain wholly minimal. Or, it may turn out that the public markers are much more affective if they have Xbees and can change your monitor’s color when you pass by them to show the travel of sound and the latent damage/impact. I’ll likely figure this out rather quickly through prototyping and user testing, and will adjust my plans accordingly.
Networked Monitoring
Look & Feel
This part involves two pieces: a network of devices that can both sense and transmit data, and a web portal that updates in semi-real-time so that users can observe the happenings from a distance. This will obviously be the most heavily technical, so it will come last in development, though it will of course benefit greatly from previous prototypes of the other stages.

This item will feature all the sensing and wireless capabilities of the aforementioned projects, but will also involve a central data uploading system so that the remote units can monitor sensor levels, report back, and send the data up to a web server. Eventually, the goal is to have the devices hidden in the environment (in my example here, Tompkins square park) and talking by Xbee radio to a similar unit with either
WiFi capabilities or a cellphone module for talking to the server (whichever is more effective for the environment.
The users will not need to know about any of this — they’ll be concerned mostly with a website that allows them to see the data in real time. Obviously, this will need to be a quick-loading site, as people won’t want to wait long for this kind of data, and will need to be simple and informative. I’m thinking the map will have circles showing (by size and color) the sensor activity detected, and get a good feel for the activity of the area.

Future Development
As this is the kind of project that would most benefit people on the go, I’ll eventually work on either a mobile application or (better yet) a mobile-optimized website so that people can plan either routes through spaces on the go. What I’m curious most about this is, provided it can be made enough available to people so that they can have reliable data to look at, whether people will use this system to track the outside world so as to plan their daily routines, whether they’ll use this as a sort of anonymized surveillance of areas, or whether they’ll find it wholly useless.

Though all of these ideas will be testing on the ITP floor before going “live” in the real world, this one stands to benefit the most from being set up and run here.
Concerns / issues to address
I’ll have a host of issues to tackle here, not least of which are the users themselves and their interest, but also the legality of the whole thing. In my experience, police don’t like circuits installed in public without permission, much less those that could be argued (somehow, however slightly) invade privacy. As such, to do any real-world tesing, I’ll probably need to get special permission and a permit, which may not be easy.
And, as I’ve hinted above, one of the biggest concerns I face is that my desire to go in so many directions may take about three times the amount of time I have left. As such, I’m starting with the lower-tech projects first, so as to have items out in people’s hands earlier rather than later — and I’m aiming to let the whole project scale as necessary. If it looks like I’m cutting it closer than I’d like, I’ll reign part of it in and revisit it on a later date. If it looks like (miraculously) I’ll have more time, I’ll allow a little more time to work on the extra elements.
Post a Comment