Reflect on your thoughts with the Commoning Interpreter
Your personal guide to community and sustainability
Imagine that you have an idea on commoning that you've wanted to share with others for a long time, or you're wondering whether your daily actions are really in the sense of community and sustainability. The “ Commoning Interpreter” is there to aide you in exactly those circumstances — a prototype that helps you find out how well your ideas and activities align with the principles of commoning.
The Commoning Interpreter is a prototype that analyses your ideas through the lens of commoning, inspired by Silke Helfrich. It shows how much your thoughts promote community, trust and sustainability.
What does the application do?
The core of this project is to introduce people to the concept of commoning. All you have to do is enter your thoughts, ideas, or activities—whether it's what you're currently doing or what you're up to. The commoning processor, powered by OpenAI's GPT-4o model technology, then analyses your input. In doing so, we have equipped the assistant with the spirit and teachings of Silke Helfrich, a thought leader in the field of commoning. The result is a reflection on how closely your ideas comply with the principles of commoning.
And what is it there for?
As in this article described, OpenAI's corpus of texts is very commercial and political. In order to communicate with this model in the spirit of commoning, we have manipulated every request to this large language model a bit by brazenly giving the agent a character who should answer in the spirit of Silke Helfrich's book “Free, Fair and Alive”:
role.user.content.prompt:"You are a scholar with extensive knowledge in the field of commoning
and embody the teachings of Silke Helfrich, as she has published in her works, including
*Free, Fair & Alive: The Insurgent Power of the Commons*. Analyze the texts in this spirit and
present them in understandable language!"The text that I am now handing over is analyzed by the system and it reflects the following idea:
role.user.content.reflektion: "Reflektiere den Inhalt, wie gut der Text die Werte des Commoning
widerspiegelt. Welche Inhalte des Textes entsprechen besonders der Logik des Commoning, und welche
widersprechen ihr besonders? ... "It is particularly exciting that this analysis is carried out in a language that is understandable to all. And that's not all: You can also enter your thoughts via voice input, which enables particularly inclusive access — no matter in which language or dialect you want to express yourself. It's a direct and intuitive way to share your thoughts.
Just try it out for yourself:
Open this application The commoning interpreter (bcommonslab.org) and also reflect on your own thoughts and activities.
Application examples
Here is an example of how to reflect on the plan of a joint picnic:
Here is an admittedly exaggerated example of how you could reflect on the personal motives of a visit to a restaurant:
Measurements values:
To make the whole thing even more tangible, the commoning interpreter also offers a type of value measurement, in which the algorithm tries to assess in percentage terms how much the content of the text corresponds to the values of the commoning. These figures show, for example, how much an idea promotes community, how trustworthy it is, or how sustainable and inclusive its impact is. Of course, these numbers are just a guide, but they give you a sense of how your ideas and actions are perceived in the context of commoning.
How does the programming technology work?
1role.user.content = {
2 "Community": "This value should express the sense of connection among people,
3 with 0 being very selfish and 100 being very community-oriented.",
4 "Trust": "This value should express the trustworthiness of the text, with
5 0 being very distrustful and 100 being very trustworthy.",
6 "Mutuality": "This value should express how inviting and open the text is for
7 collaboration, with 0 being very dismissive and 100 being very inviting."
8 "Sustainable": "This value should express how consciously resources are managed,
9 with 0 being very wasteful and 100 being very mindful and frugal.",
10 "Inclusive": "This value should express how inclusive the text is, with 0
11 excluding certain people and 100 including everyone.",
12 "Social interaction": "This value should express how much the text promotes
13 collaboration and the fostering of relationships, with 0 indicating antisocial
14 behavior and 100 indicating highly social behavior.",
15 "Equal self-organization": "This value should express how much the text
16 encourages negotiating on equal terms, with 0 being very hierarchical and 100
17 promoting encounters on an equal footing.",
18 "Caring, self-determined economy": "This value should express how much the
19 text reflects caring and self-determined economic practices, with 0 being very
20 externally controlled and profit-oriented, and 100 being self-determined and
21 needs-oriented."
22 }
The generated values should of course be treated with caution, but show how large language models can be used to express perceived orders of magnitude in numbers.
Open source
If you are interested in taking a closer look at how the application works and tinkering with it yourself, this option is also available:
bcommonslab/chatWithFreiFairalive (github.com)
You can try out the chatbot here:
Chatbot (bcommonslab.org)
Call to Action
The project is still in the prototype stage, and this is where your chance comes in. We are looking for people who want to get involved and further develop the project. It is particularly important to build up your own corpus of texts with scientists in order to make the analysis even more precise and trustworthy. So far, the analysis has been based on information from the Internet, which does not always offer the most reliable sources and is extremely commercial-oriented. But that is exactly what the potential shows: If we use artificial intelligence wisely, we can create valuable support for human decision-making processes.
