How Urban Farmers Learned to Stop Worrying and Love Big Data

Hello everyone, and welcome to week five! My name is Ricky Mouser and I’m on the Truly Living Well team. As we near the halfway point of the DSSG program, it seems timely to review the work we have put in so far, while keeping an eye out for what is still to come.

That’s me on the right, hearing from farmers as we pick spinach. Credit: Laura Diamond
That’s me on the right, hearing from farmers as we pick spinach.
Credit: Laura Diamond

Urban agriculture is a complex undertaking with many inputs and outputs to manage. Even apparently straightforward operational information such as visitor demographics, timecards, or harvest requests can be difficult or impossible to gauge without a streamlined data management system. Therefore, we need to determine how to satisfy the data-driven needs of support staff without unduly burdening farmers with the onus of collecting this information. If data collection seems more trouble than it is worth to the actual farmers on the ground, it will likely be left by the wayside, and TLW will be no better off for our efforts.

Initially we believed that the data management problems faced by a farm could be handled by some pre-existing website or app that applied established best practices to the solution. But while these tools do exist, they are invariably tailored for commercial farming, which is qualitatively quite distinct from urban agriculture. With the scale of commercial agriculture, most labor is best supplanted by automation at these scales, leading to inadequate worker tracking. Additionally, optimization of crop output becomes trivial, and planting whole fields of just one or two crops can be a viable solution. This culture of bland mechanical optimization is infectious, and it quickly became clear to us as a group that these “big-ag” solutions would not accommodate the work of TLW, a group of small, volunteer-run farms attempting to supply underserved communities with most of their diverse produce needs. We needed to create a totally new, custom solution of our own.

At first we thought that we would be responsible for gathering all data ourselves, but a bit of digging has shown that by integrating across pre-existing databases maintained by the POS system and Quickbooks we can minimize duplication of collection efforts. While we were glad to see that digitally tracking farm output is not entirely foreign to TLW, we realized that needed to interview as many farmers and technical staff members as possible to assess what additional metrics they actually needed to track to better understand their operations. At the suggestion of Dr. Carl DiSalvo, we used Balsamiq, an easy-to-use piece of software that allowed us to iterate through several mockups of a data collection website in days instead of weeks.

These meetings were very rewarding for both sides. For us, the iterated feedback we have received makes us confident that the final product we deliver will be well-suited to the needs and technical abilities of Truly Living Well. But even more rewarding was seeing the shift in attitudes of both farmers and staff as they saw our prototypes coming into alignment with their vision. Moving away from pencil and paper to entering standardized data into a tablet can be daunting for farmers accustomed to working with their own hands and doing things the same way they have always done them. But as the benefits of having clear, accessible data became obvious to them, we got to see these volunteers and workers light up one by one with the realization that while change is always difficult, a new, simpler system has the potential to make their lives tremendously easier and their operation more adaptive.

After noting feedback from various staff members, we are now in the process of creating a minimally functional prototype of the system itself. On the technical side, the key for us going forward will be figuring out how to merge these databases with our own in a way that is maximally flexible and easy to use. As such, we are designing our website with the simplest tools possible: HTML, Javascript, and MYSQL for the database management. Hopefully this will allow us to create a system tailored to meet exactly what TLW needs, while remaining simple to allow for any future extensions of functionality to be incorporated as easily as possible.

We’re excited about our progress, but even more eager to get started on the second half of the program. Cheers!