The same, however, is not yet true for making product purchasing choices based on environmental impact. While some eco or green labels have started to emerge (e.g., EnergyStar in the US), we generally don't have the information available that would allow us to act on whatever our environmental convictions may be. There are many dimensions to the possible environmental impact of a product, such as whether the materials it is made from have been or can be recycled. Another important aspect that a lot of people care about are the emissions that a product caused during its manufacturing process and will cause during its operation. That turns out to be a much harder measurement problem to solve than the materials problem.
Consider a new computer. When the computer runs, it consumes electricity. The emissions caused by operating the computer therefore depend not just on how much you use it but also where your electricity comes from. Then there are the emissions that were caused by getting the computer shipped to you. That introduces more difficulty. How far did the computer travel? Was it shipped by truck the entire way or part of the way by air? What kind of truck was it? What type of airplane? But if you crack open the computer, then it really gets tricky. Every part that goes into the computer has its own emissions history. It too was manufactured and then transported to get to the place that assembles the computer. Determining emissions is thus a massive undertaking of connecting activities with their emission factors (which may well vary even for the same activity depending where or when it is carried out).
It is at this point that most people just give up and declare this problem as intractable. The team at AMEE, instead saw an opportunity for a lightweight web service. AMEE is a database in the cloud that allows tracking of activities and applying emission factors data to the tracked activities. It was built from day one to support other applications and web sites which gather the activity data from consumers and businesses. For consumers, that includes carbon calculators, e.g. Google UK's Carbon Footprint Project, but also services such as DOPPLR which uses AMEE to automatically calculate emissions for any trip it tracks. For businesses, existing systems that can plug into AMEE include accounting and supply chain applications.
The beauty of the AMEE approach is that instead of trying to solve the entire measurement problem in one go, AMEE starts with a known level of detail even if at that level of detail only estimated emissions factors are available. For instance, AMEE might have an estimate of the energy embedded in computer based on weight and materials. AMEE might also have an estimate of the laptop's power consumption based on average usage. And so on. As more and more systems connect to AMEE and provide data, those initial estimates can be refined and triangulated.
As Tim O'Reilly has described in his "Web Meets World" theme, we are on a path to increasingly instrument the world. Every time a new measurement system comes online, whether it is smart meters in homes, or cars that talk to the network, the data can be fed into AMEE allowing for a gradual transition from estimated emissions to detailed actual measurement of emissions. AMEE has made the deliberate choice to focus only on acting as a backend database and web service so as to not compete with any of the systems and companies that could provide inputs for AMEE and thus help improve the quality of AMEE's data asset.
All of this information will in turn let consumers and businesses make purchasing and production decisions that take environmental impact into account. We firmly believe that making more information available will ultimately result in better decisions and that the Web is the perfect mechanism for empowering people to do so. We are therefore excited to be investing in AMEE together with O'Reilly AlphaTech Ventures and TAG and look forward to working with Gavin Starks and the AMEE team.
- albertwenger posted this.