The goal of this hackathon is to help "fix the soil health tech stack" by writing code together that both produces and uses soils data, thereby demonstrating interoperability and improving toolsets available to the digital soil community specifically around soil sample results and related information.
This is different than a normal hackathon where teams compete to out-do one another in particular challenges. This hackathon is about shared engineering collaboration: we are all working together toward the final demo, and the more that works the better it is for all of us.
The Fixing the Soil Health Tech Stack hackathon will start at 9:00 am Central Time on August 23 and conclude with final demos the next day starting at 12:35 pm Central Time on August 24.
Participants can decide which teams should form in order to achieve the shared goal. In the leadup to the hackathon, we will use the "fixingsoil" Slack organization to decide on teams and shared final demo goals so that heading into the hackathon we will have a clear runway to getting things working. All registered participants for the hackathon will receive an invite to the slack team.
Once at least a first draft of available teams has been decided, we will update this page with the teams list and proposed final demos.
The hackathon will take place in the virtual events platform Remo.co. The main "virtual floor" is a set of tables that participants can move freely between as they have discussions. We intend that each team will have a dedicated table, and any participants can move between teams to collaborate or ask questions. It is also easy to have ad-hoc small discussions by just having a few people pop over to an unoccupied table. The "virtual floor" for this hackathon will look like this:
Each team will elect a "speaker" who will be in charge of handling communicating the teams progress during the hackathon and presenting their part of the final demo. A team may have more than one person speak at the final demo, but one person should be in charge of communication of this sort.
You can find the schedule at the Event Hub in Swoogo as well as all the links to the video conferencing platform(s): https://farmfoundation.swoogo.com/soilhealthtech/sign-in?rId=1858626
The most important times: Tue Aug 23 @ 9:00 am Central: Hackathon Kick-Off Tue Aug 23 @ 11:30 am Central: Demo planning - make sure you are here if you want to include your work in the final demo! Wed Aug 24 @ 11:35 am Central: Aaron will check-in with all teams to finalize demo and who is speaking. Wed Aug 24 @ 12:35 pm Central: Present Hackathon Results at main conference Wed Aug 24 @ 1:10 pm Central: Hackathon is done!!
The goal of the final demo is demonstrating interoperability between tools regarding soils data. The tools, libraries, apps, etc. developed at this hackathon will likely fall into one of these categories:
- Form: Models of soil sampling data
- Backfill: Populating the standardized models with past data of various forms, or exporting the standardized forms back into CSV and other formats.
- Function: Apps and Analysis that can consume soil samples in the standardized models and show them in a UI or chart.
We have provided several examples of soils data, housed on github (and importable via npm in javascript) in the examples/ repo under the 'https://github.com/oats-center/modus/' monorepo.
git clone [email protected]:oats-center/modus.git
cd modus/examples/examplesList of example types: https://github.com/oats-center/modus/tree/main/examples/examples
In general, each example folder contains one or more "source data" sheets in XML, CSV, or XLSX form, and then the @modusjs/cli command-line client was run on the source data to produce Modus JSON-formatted json files. During the hackathon, we hope to add an exporter from the json files to a standardized CSV/xlsx structure. Use whichever versions fit your situation.
Thank to the Skidmore College team for collecting several hundred soil sample cores. They collected cores from particular locations and sent soil from the same sample to multiple labs in order to compare the Total Carbon measurements across labs. This lab data has been anonymized and provided here. In addition to Total Carbon, some of the cores also were sent for a soil moisture measurement as well. Both types of data are included in the lab results.
Thanks to Point Blue Conservation Science (https://pointblue.org) and TomKat Ranch (https://tomkatranch.org/) for providing several years of historic soil sample data! The tomkat_source_data.xlsx file here contains many sheets with historic data for the same locations. Also, there are CSV files here representing the lab results structure reported by Ward labs. Both are convertable using the @modusjs/cli command-line client.
We have run the client on all the source data and produced all the .json Modus JSON files included in this folder. Please use whichever version fits your situation.
This folder is a single soil sample from Ault Farms, Inc on the "Enyart East 50" field, sent to A&L Labs Great Lakes. It includes:
- result.pdf: a PDF report that typically is sent to the farmer,
- lab-modus.xml: a Modus XML version created by the lab
- hand-modus.xml: a hand-created Modus XML version derived solely from the PDF file,
- hand-modus.json: a hand-created Modus JSON derived from the hand-created XML file.
These two XML file are the only original Modus XML examples we have at the moment, so please use those if you would like to demonstrate supporting Modus XML. If you are looking to understand the JSON structure of Modus, a great place to start would be looking at the PDF and looking at the hand-modus.json file side-by-side to see how they match up.
We have chosen the Modus format hosted by Ag Gateway as the main soil sampling result model for this hackathon. We've made a JSON schema which faithfully represents the original Modus XML spec, a converter for XML into that JSON schema. We hope to also develop, either prior to or at the hackathon, a "slim" version that is simple to navigate.
These can be found at the Modus Bitbucket repo here: https://bitbucket.org/modus/modus-schema/src/Version-1.0/.
There are three components of the JSON version of Modus (to mirror the original XML structure): global, modus-submit, and modus-result. modus-result uses definitions found in global and is the form in which a soils lab would communicate lab results. Hence it will be the focus of this hackathon.
These three components are published as JSON schemas (compiled from Typescript), and associated Typescript types.
-
globalSource (TS): https://github.com/OADA/formats/blob/master/schemas/modus/v1/global.schema.cts JSON Schema: https://formats.openag.io/modus/v1/global.schema.json Typescript Types (npm): `@oada/types/modus/v1/global' -
modus-resultSource (TS): https://github.com/OADA/formats/blob/master/schemas/modus/v1/modus-result.schema.cts JSON Schema: https://formats.openag.io/modus/v1/modus-result.schema.json Typescript Types (npm): `@oada/types/modus/v1/modus-result'
A monorepo of tooling is avaiable here: https://github.com/oats-center/modus.
examples: @modusjs/examples. Directly import-able XML and json examples of modus (great for testing).convert: @modusjs/convert. Javascript library to convert between formats. Currently supports converting a Modus XML string into Modus JSON.cli: @modusjs/cli. Command-line tool to perform file conversions (i.e. xml to json).
Thanks to all who have elected to participate in this hackathon and we look forward to the great thing we can do together to "Fix the Soil Health Tech Stack!"