Using the methods of critical spatial practice, I wish to examine the conditions under which building owners are likely to install a fire sprinkler system. To facilitate the data collection process, I created an iPhone application called PhotoLocatr that geolocates and tags photos taken with a digital camera.

building code

New York City building code requires buildings to have signs that indicate to firefighters parameters of a standpipe or fire sprinkler system’s operation. The standard is set forth by a national rule-making body, the NFPA, in NFPA standard 14. The standard reads:

6.4.5.2.1 If automatic sprinklers are also supplied by the fire department connection, the sign or combination of signs shall indicate both designated services (e.g. “STANDPIPE AND AUTOSPKR,” or “AUTOSPKR AND STANDPIPE”).

You and I can see this standard manifested as signs on buildings across the nation. Standpipe/sprinkler protected buildings can fall into three main classifications: fully sprinklered, partially sprinklered, or not sprinklered.

Colored covers (or paint) on the standpipe indicate how the system works; red means the hookup connects to a standpipe only (i.e. does not feed sprinklers, or none are present); green means an automatic sprinkler system is present and connected to the hookup; aluminum color means sprinklers are present but not automatic (requires turn-on by the fire department), and yellow signifies a combination sprinkler/standpipe system (both are fed by a common standpipe).

See this document for more information about NYFD standpipe standards and usage.

data collection

My research process consisted of collecting geolocated standpipe photos in New York City (specifically Manhattan for now), and mapping the results by one of many dimensions (e.g. full, partial or no sprinkler coverage). To do this, I created a tool called PhotoLocatr that runs on the iPhone and makes use of Flickr to store and manage the photos.

Part of the collected data appears below:

analysis and conclusions

Raw data collected so far appears above (view on geocommons); orange dots represent “full coverage” by sprinklers, while grey dots represent partial or no coverage.

General observations include:

• Lobbies and halls are often protected, even if the entire building isn’t. This gives occupants more of a chance to escape a fire, but without the expense of adding sprinklers to every office or room.
• Compactor rooms and or garages are sometimes the only places protected by fire sprinklers.
• Older buildings usually have standpipes, but no sprinklers–not surprising, as installing sprinklers (and the necessary pipe) in an old building can be disruptive and prohibitively expensive, especially if one must do so in each of hundreds of rooms/offices.
• Most newer buildings, or buildings frequently renovated (think Midtown) often have sprinkler systems–again, making significant renovations to a building forces one to comply with modern building code, so no big surprise here.
• Every New York Public Library I passed had full sprinkler coverage.
• St. Vincents, a local hospital, has complete sprinkler coverage.
• MTA subway tunnels have standpipe hookups at the street level for firefighter use underground.
• ConEd installs sprinklers in and around their electrical transformers, making it easier for firefighters to extinguish electrical fires quickly and safely.

Update! See my article on the Infrastructurist for a summary of my findings around this project!

ConEd asks the people of New York to report steam they see on the streets of lower Manhattan. ConEd provides a phone number (1-800-752-6633) for the purpose, but Steamr hopes to make the reporting process faster, easier and more accurate using the advanced capabilities of the iPhone.

It is hoped that this project benefits both the public and ConEd alike, setting an example of how the public can help utilities maintain our increasingly aging infrastructure, and starting a dialog around public interest and concerns of aging infrastructure.

The project was conceived during my time as an artist-in-residence at The Banff Centre New Media Institute. Special thanks also go to Alexandra Woolsey Puffer for lending her graphic design skills to the effort!

presentations

This project was presented as part of Banff’s “open spaces day”:

Neighborhood Networks is a joint project of the CREATE Lab and the University of Pittsburgh Center for Learning in Out of School Environments (UPCLOSE), and is funded through a grant from the Intel Corporation.

As part of my involvement with the project, I rode my bike roughly 40 miles over two days in Pittsburgh, Pennsylvania with industrial hygiene sensors. I used the Casella 360 noise dosimeter to measure and log sound levels, and the Vulcain SafetyPalm IAQ Monitor to measure and log carbon monoxide and carbon dioxide levels. GPS data was recorded with the Garmin Forerunner 201. The former two were rented from Field Environmental in Pittsburgh.

data collection

In the first ride, I put the sound dosimeter in my backpack, and clipped the microphone sensor to the back flap of my bag. The gas probe was put in the waterbottle pouch (mesh) of my backpack. I rode about 20 miles on this trip, at a normal speed. Results were good overall, but gas levels seemed to be uniformly low (and this seemed wrong)

For the second ride, I made a mount of cardboard to keep the gas probe out in the open–it would not touch any surface during use, and would have free-flowing air on the sides (where the sensors are; except for the stalk, which measures temperature). I also tried to go no more than 20 mph during the entire ride. This configuration seemed to work much better. I kept the sound dosimeter in the same place; it seemed to work well the first time. This trip was also around 20 miles long.

analysis and conclusions

I imported the logged results from each of the sensors, ran them through some custom scripts I wrote (available below) and plotted the resulting KML on Google Earth. Everything used, along with the output of the scripts (KML) is available below.

sensor ride 1 data
sensor ride 2 data
KML generator script
Cel-360 config file used

The results painted an interesting picture of the city, capturing some local ephemera: the Kenny Chesney concert, and the necessary traffic bringing people to it. In the screenshot of Google Earth shown above, you can see a large spike in carbon dioxide levels when I passed under the I-279/PA-65 overpass. I can only assume this overpass traps gas below it (it wasn’t windy on the day I rode), and the sensors picked it up. Also notice the increased levels along the riverfront: there were large numbers of idling boats and generators powering TV’s, radios and barbecues–again, all for the concert.

I think a “sensor ride” is an interesting way to experience each neighborhood in Pittsburgh (I passed through at least a dozen), and a more visual way to look at the issue of pollution: both gaseous and aural. Next steps might include visualizing the data by zoning classification or socioeconomic background, or including a personal narrative of the ride (i.e. “there’s a bus next to me now”) as part of the data.

Research methods ranged from collecting and cataloging litter, to attending church dinners as participant-observers in neighborhood life. The project resulted in us leading hopeful, creative workshops for youth to express their ideas for improving the neighborhood, which culminated in postcards sent to the Mayor of Pittsburgh.

We collaborated with a local design and strategy firm to conduct ethnographic and design research, and to create the information architecture and visual design for the website. The website’s significant new features include an expanded citation database, more complete search functionality, interviews and other original content.

The website is a project of the University of Pittsburgh Center for Learning in Out of School Environments (UPCLOSE), and is funded through a grant from the National Science Foundation.

The Global Connection Project hired DeepLocal to develop a website that would serve as the public outlet to share these special high-resolution images. We met with stakeholders to understand their vision and research interests, and then facilitated a participatory design workshop to develop a human-centered solution. With the insights from the workshop, I led the engineering of the technology.

Gigapan is part of the Global Connection Project, a project of Carnegie Mellon University and NASA’s Ames Research Center.

MapHub began as a project of The Carbon Defense League at The Studio for Creative Inquiry. The team worked with local community groups to position MapHub as a new channel of communication between city government and residents. The MapHub solution invited community members to enter information with its simple interface, while transparently allowing public officials access to the data.

Because of its potential value to so many audiences, the University spun-off the project to form DeepLocal. My involvement with MapHub spanned three years, through the first four revisions of the software.

Myself and the other two co-founders were among the “Top 40 Under 40” entrepreneurs recognized by American Venture Magazine in 2007.

Participants in Pittsburgh made recordings around the city using their mobile phones. Participants input the location of the recording by GPS coordinate, street address, or intersection. The recording was then uploaded to the MapHub system, accessible for playback or for commenting on the web.

For the physical device installed in Karlsruhe, we collaborated with artist Greg Baltus to create a robotic mechanism that played the recordings in real time. It moved an LED marker behind a translucent, wall-mounted map to the location of each recording as it was played (see video).

The exhibition was curated by Peter Weibel, Bruno Latour, and Steve Dietz, and a companion anthology (also titled Making Things Public: Atmospheres of Democracy) was published by the MIT Press in September 2005.

“Agents” navigate through the city, visiting various locations to solve a topically relevant “mystery” in the city of Pittsburgh. The experience was augmented by technology including tablet PCs, barcode scanners, RFIDs, GPS, digital cameras and cell phones. My involvement with the project spanned two years, from research and development (including four extensively tested iterations) to implementation and eventually summative research.

Click! was jointly developed by the University of Pittsburgh Center for Learning in Out of School Environments (UPCLOSE) and the Carnegie Mellon School of Design. The program is now owned and operated by the Carnegie Science Center in Pittsburgh.