Great presentation by Saul Griffith at ETech this year. Two points stick. 25min 25sec in he compares his "low estimate" of his actual energy usage and it turns out to be much higher than all other footprints currently being used on the internet. The slide after then makes the point beautifully by suggesting the next Nobel prize for economics should be awared to the person able to accurately calculate our energy usage... The second point was 36min 40sec in. The list of what we as individuals can do was brilliant "these are all things we want to do anyway" - spot on. So why don't we do it.

I admire his goals for 2010 and really hope he obtains them.

I have been asked by a few people to provide a summary of the project and what we learnt. A summary pdf was created for the final conference. Below are my thoughts.

What did we do? BOP gathered quantitative and qualitative data about the physical environment, the use of space and the mood of the work force. The WSN toolkit enabled the collecting, manipulating and displaying of both tangible environmental factors, for example, light levels, heat levels, noise levels and people’s presence, and workforce reports on intangible factors, such as perceptions of personal energy levels, sense of well-being, stress and feelings of connectedness with others. In practical terms this meant we deployed a 20 node WSN capturing environment data, activity based sensors and prototyped several different polling devices.

Crossbow WSN - light, temperature, pressure

Arduino based CO2 sensor

Arduino based activity sensors - presence at meeting tables

Beastie based activity sensors - ultrasounder to monitor corridor activity

Weigh your opinion - polling devices

Visualisation - ticker playing in entrance to office

Visualisation - intranet, screen based

Auralisation

Why do this? Organisations are potentially interested in space from three main perspectives: cost efficiency; employee performance; and brand image. Current measurement practice relies largely on hard-wired sensors, for monitoring of building services, and on manual clipboard surveys and/or online surveys for occupant feedback. Relative to these approaches, the toolkit offered several advantages, such as easier, cheaper data collection, more rapid analysis/presentation and the wherewithal to collect a broader set of data. In addition, the use of playful ‘front-ends’ offered the prospects of higher participation rates, hence more complete data sets, while the use of rapid visualisation techniques offered the prospect of a speedy feedback loop to building/office managers, line managers or the work groups themselves.

What did we learn? Here are my top three:
- The WSN technology worked but takes longer than acceptable to set up and use. Getting a network configured to reliably measure the data you want without intervention is not trivial.
- It is difficult to solicit qualitative environment information in an automated manner. Whilst the polling devices worked well I often found the data analysis sessions left me wishing we had measured more, or differently. I often doubted the scientific validity of the conclusions we reached.
- Presenting the sensory objects and data in a human readable manner increased peoples desire to participate / interact with an object. Sensing and monitoring should be a two way street. (maybe this is why self surveillance works)

Ubiquitous computing is still in its commercial infancy. BOP was the first project to use WSN to get a better understanding of the fit between built environments and the organisations and people who use them.

Other resources:
more photos on flickr
summary podcast with ppt slides
video of an IET presentation I made on this project