MIT “Food Computers” Project

The MIT Media Lab has a flashy project called the Open Agriculture Initiative they call “food computers”, which appear to basically be heavily automated greenhouses.

The project is addressing the challenge of food production, particular for dense, poor, urban populations. The vision is for everyone to grow their own food locally, cheaply, efficiently, and sustainably.

The general idea is an open source set of hardware, software, data, and knowledge that enables people to build their own “food computers” at different scales, from small “personal food computer”, to “food server”, and ultimately “food data center”. (These terms are adapted from computing, they deserve better names.)

These “computers” are a “controlled-environment agriculture technology platform that uses robotic systems to control and monitor climate, energy, and plant growth inside of a specialized growing chamber.” Food production is optimized by monitoring and controlling “variables”, such as “carbon dioxide, air temperature, humidity, dissolved oxygen, potential hydrogen, electrical conductivity, and root-zone temperature”, as well as optimizing energy, water, and other inputs.

One of their innovations is organizing these monitoring and control regimes into what they term “climate recipes”.  These are basically scripts that make it easy to reproduce specific growth patterns. These recipes are portable and potentially could be up- and downloaded, shared, and mashed up. (Here they are emulating the successful model of digital file sharing and personal fabrication.)

It seems to be early days on this project. There are some alpha level downloads, mainly CAD plans to make the personal FC hardware and related software. I don’t see much information about the recipes, which I image are a programming language to describe food growing. That sounds kind of interesting, as will be the ‘high level’ software that creates and optimizes scripts. For example, I could imagine a tool that takes in planning data about resources and population to serve, and generates ‘optimized’ networks of FC’s (so many large servers, located here, so many personal FCs, located there, timelines for upgrades, seasonal adjustments for available water and power, and so on.)

There are lots of cool ideas here, even if the terminology is terrible. I understand why you don’t want to call these things ‘food factories” or even “food fabricators”, and “robot green houses” isn’t all that great. But “food computer” is just plain misleading, and the term “food data center” is absurd. Let’s get some better names here. “Personal Cornucopia?”  Shorten to ‘my copia’?

I have to say, though, that I’m not entirely enthusiastic about this project. There is much to say for local food production, and if people live in a city and there is no better choice, then something like these FC’s is better than starving.

The project argues that this approach is “sustainable”, in the sense that it is vastly less wasteful than conventional agriculture, and produces a lot more food for the same inputs. This is probably true (when he technology is working correctly), but I have to think that it is beside point. Much of the waste in agriculture is not due to ignorance, it is due to economic incentives. Even if every poor person grows their own veggies, industrial agriculture will still raise vast herds of cattle and feed grains, “mining” the planet to sell for cash from the wealthy. Just like today.

As a long time gardener, I have to worry about their extremely non-biomimetic approach. These robotic environments are purely artificial environments, designed to be better than nature. The machine “knows” about a handful of variables and has a handful of knobs to turn. The system itself is neither self regulating not self regenerating, and definitely not “sustainable” without human intervention and inputs. This is the opposite of biologically inspired design, it is anti-nature.

The idea of “recipes” is cool. It is, of course, formalizing what gardeners do in their heads informally. (We’ve been exchanging gardening knowledge since the invention of language—it’s probably one of the first things humans ever talked to each other about.) As a veteran computer programmer, I have to worry about such software: scripts are fragile, stupid, and non-adaptive. (They can also be hacked.) I’d hate to see millions of people dependent on a software monoculture, that could crash due to some bug or inability to handle a new situation.

Most importantly, I have several problems with the psychology of this technology.

One of the great benefits of personal gardening, in my opinion, is not the harvest of food or flowers, it is the process of working with nature to bring forth life. We learn to think long term (whole seasons, months long, “next year”), we learn think about the big picture (we are happy to see snow in the winter, we rejoice at the first bee of the spring, we feel the warming and then cooling of the sun as the season passes). We touch dirt, and bugs, and all kinds of living and dead matter. It’s real.

Above all, we learn humility. Even a small garden is a wonderful thing, something we can guide but never really control. We have a better idea of where we fit in the universe. These attitudes lead us to be grateful for our food, whether we grow it ourselves or not, and to care about the planet in a more than theoretical way.

The “Food Computers”, on the other hand, teach us that food is a commodity that we need to make as efficiently as possible, just like cars or shoes. The human sets up the system, loads the program, and pushes the button to start. The human cannot sense what is going on, and has the same relationship with the robot garden as with the thermostat on the furnace.

Life is a spreadsheet of inputs and outputs. It’s small scale thinking (adjust this sensor reading, adjust this input valve), short term thinking (how fast can I get something out?), and the whole thing is about human control.

We see the latter in the hubris of such statements as “Imagine growing water-loving tropical fruits in the middle of the desert, or sun-loving summer berries in the midst of a snowy Boston winter. With climate recipes, you can grow local from anywhere!”  An engineer might imagine this, but would a gardener?

For these reasons, I would expect that anyone who has a choice will not choose to use these “food computers”. Maybe we will have no choice. But if we do, we’ll garden for pleasure the old fashioned way, and we’ll try to grow food more naturally if we can.


(These computers also have a touch screen, so they might be candidates for the Inappropriate Touch Screen Files.  For now, I rule that, however flawed the concept, the touch screen interface itself is not egregious.)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s