Updated: Sep 7
We harness the web for life cycle assessment (LCA) and supply chain sustainability tracing.
Can you believe that Earthster, a platform to harness the web for distributed life cycle assessment (LCA) and supply chain sustainability tracing already existed 15 years ago? With a tiny band of talented and committed developers and designers, our alpha got a few people and companies very excited. But lacking startup experience I took an academic consortium path which end up abandoning the transformative scaling vision that had made Earthster exciting in the first place.
Six months ago a talented pair of gents rich in startup experience, investor savvy, and even complimentary LCA expertise found me, and I told them “you’re going to need a tool to do what you’re proposing.” They agreed and asked if I had any ideas, and here we are: fresh talent, a decade of learning, and an amazing core of passionate know-how bringing an updated Earthster to life the way it was always meant to be.
“Earthster is the economy coming to know itself.” I still love this original, poetic expression of the Earthster vision. But I find that a more accurate and useful description is:
Earthster is the actors in the economy coming to know and to continually improve their relationship with the earth.
To enable this global activity Earthster uses the language, data, and methods of Life Cycle Assessment (LCA) to describe the relationship between an economic activity and the earth.
First, there are the flows of pollution and wastes released from the activity into the earth’s air, waters, and soil. This includes flows like greenhouse gases released to the air, carcinogens deposited in soils, and nutrients discharged to water. Second, there are flows of resources from nature into the economy; here we mean flows like water, timber, and ores. Third, there are flows of goods and services that connect the economic activities to one another, forming supply chains and product life cycles that can span the earth itself. These economic flows mean that a decision or action taken somewhere on earth will have consequences that ripple out to alter the environmental inflows and outflows occurring at thousands of other economic activities.
Earthster uses these quantitative LCA pictures of supply chains and life cycles and their interactions with the earth so that decision makers sitting somewhere (anywhere) in the economy can make local choices that ripple out to be better for the environment across the life cycle and around the earth. These pictures also enable other actors to reward and encourage environmentally beneficial choices by others.
The big question for today is: how can Earthster obtain at scale, so that it can offer at scale, these LCA pictures of the environmental and economic interactions of activities and supply chains and life cycles? How do we fulfill the need for data-driven models of the life cycles of millions of goods and services?
Earthster is coming to life in a world where the LCA data problem is still being solved the same way it was solved when the first LCAs were done back in the late 1960s. It’s the approach we might call the “omniscient modeler” by analogy to the concept of the “omniscient observer” who has very great or seemingly unlimited knowledge.
In this approach, which is quite common in the building of models in science and engineering, an expert or small team of experts collects a huge amount of data to build comprehensively descriptive models of the environmental impacts of the main activities across many of the life cycles and supply chains most likely to be studied. Collectively these models comprise what are called “background databases.”
Then, in order to conduct a particular study, a modeler leads or directs the gathering of “foreground data” about novel, new, or local parts of a life cycle, and connects it to the models from the background database.
The number of model builders is small (a few thousand), and the number of database builders is even smaller (from a handful to a few dozen). Some of the most economically successful and powerful companies in this space have, from the start, merged the activities of modeling and database provision, deploying a business model that allows them to feed foreground data that they collect into their proprietary background database.
Earthster introduces an approach that starts with the conventional LCA data world as its foundation, and then introduces user-driven data generation on top of it. We have set out to empower all actors in the economy to make LCA-informed decisions. If we can empower millions of users, we will have simultaneously empowered and incentivized millions of LCA data collectors.
To build LCA models of their products, these actors will use their ready access to information about their own process’s interactions with the environment and with the economy. At first they will do like today’s niche of a few thousand specialized model builders: they’ll link their foreground data to some background data in order to obtain and use their own model of their own products.
A popular open source maxim is ‘With enough eyeballs, all bugs are shallow”, meaning that projects with lots of contributors have higher bug fix rates. Earthster’s corollary is: with enough LCA modelers, all data is foreground.
Earthster sets up a situation akin to the famous story of the blind people and the elephant. In this fable, a group of blind people encounter an elephant, and while each has valuable local knowledge, none is able to perceive the whole. A person at a leg shouts “it’s a tree”; a person near the elephant’s broad side shouts “it’s a wall.” A person at the tail shouts “it’s a rope” while a person at the trunk shouts “it’s a snake.”
How do they arrive at the full and accurate picture of reality? They can’t take a vote or they would likely conclude it’s a tree. They can’t adopt the view of the best arguer or the most senior. The accurate picture comes from a combination of (a) listening to everyone, and then (b) collectively creating a synthesis, leading to more than the simple aggregation of what each local observer thought they had found.
In our case, the blind people are the actors in the economy. And the elephant is the economy’s relationship with its natural/physical/living home. Earthster allows unit processes around the world to express themselves, and it creatively synthesizes the data from each, fed back to all.
The omniscient observer approach with a small band of specialists works OK when the reality being modeled is changing slowly enough. In a slow world, we can wait for the small band of specialists to make their way around the cathedral, measuring and recording one brick at a time. But our world is more like a walking elephant, and one that now needs to break into a sprint.