Difference between revisions of "Stigmergy"
| Line 25: | Line 25: | ||
'''At the second level''', we have '''active signaling stigmergy''', which means that agents can get notifications from ques that are deposited in other contexts, based on some criteria. This second level was implemented in [[Sensorica]]'s web2 version of the [[NRP-CAS]]. For example, if an [[affiliate]] has performed tasks in a [[context of work]] (i.e. a [[project]] or a [[venture]]), whenever there's another task of the same type that is created by another [[affiliate]], the first [[affiliate]] can get an email notification. This allows a group of affiliates in a smaller size network perform well, by increasing the probability of an action being taken (someone will take a new task) in a different [[context of work]], relaxing the requirement of the critical mass (high number of [[agents]]). This is a dumb as a matchmaking system. The idea is to reduce the distance / proximity requirement. | '''At the second level''', we have '''active signaling stigmergy''', which means that agents can get notifications from ques that are deposited in other contexts, based on some criteria. This second level was implemented in [[Sensorica]]'s web2 version of the [[NRP-CAS]]. For example, if an [[affiliate]] has performed tasks in a [[context of work]] (i.e. a [[project]] or a [[venture]]), whenever there's another task of the same type that is created by another [[affiliate]], the first [[affiliate]] can get an email notification. This allows a group of affiliates in a smaller size network perform well, by increasing the probability of an action being taken (someone will take a new task) in a different [[context of work]], relaxing the requirement of the critical mass (high number of [[agents]]). This is a dumb as a matchmaking system. The idea is to reduce the distance / proximity requirement. | ||
'''At the third level''', we have what [[Tibi]] calls '''proactive stigmergy'''. We can also understand this as part of the [[feedback system]]. One way to implement this is to make the collaborative digital environment smarter, using machine learning or AI agents. In other words, AI agents can scan activity and flows within the environment and detect priorities in some contexts of work and send signals / ques / notifications to specific agents who can, directly or indirectly intervene. At this level we increase the general awareness about important issues throughout the network and further reduce the requirement for a critical number, allowing even smaller networks to operate well. | '''At the third level''', we have what [[Tibi]] calls '''proactive stigmergy'''. We can also understand this as part of the [[feedback system]]. One way to implement this is to make the collaborative digital environment smarter, using machine learning or AI agents. In other words, AI agents can scan activity and flows within the environment and detect priorities in some contexts of work and send signals / ques / notifications to specific agents who can, directly or indirectly intervene. At this level we increase the general awareness about important issues throughout the network and further reduce the requirement for a critical number, allowing even smaller networks to operate well. | ||
| + | |||
| + | =Stigmergy and Governance= | ||
| + | |||
| + | Notes from meeting with the Metagovernance project and TaoDao. In the room: Tibi, Cent (Metagovernance), Daniel (Active inference lab), Ronen (metagovernance), Vini (TaoDAO), Tim. | ||
| + | |||
| + | Exploring governance and stigmergy, how these two conjugate together. | ||
| + | Commons sense makers, stigmergy markings, open source attention. | ||
| + | |||
| + | In a social dynamic heavily driven by stigmergy rules are important but group decisions are disappearing. | ||
| + | |||
| + | We are exploring the interplay between governance and stigmergy, | ||
| + | |||
| + | Can we create a stigmergy / governance workshop? | ||
| + | Vini says that people need to be able to reframe their attention and interest, in order to be able to sense their environment, the ecosystem. One requires an enhanced perception of the environment. | ||
| + | |||
| + | Set up a game field, see who picks it up, and those who come to the workshop, will be already selected by their awareness. | ||
| + | Give time to incremental amounts of people. | ||
| + | It can be an asynchronous experience. | ||
| + | Make it easy to engage. | ||
| + | Necessary to move the attention towards the ecosystem. | ||
| + | |||
| + | Stigmergy is like governance focused on attention. | ||
| + | Governing our collective attention. | ||
| + | |||
| + | For stigmergy to work there is the network and the environment, and pressure from the environment that drives self-organization and optimization through stigmergy. There is a pressure for selection on the group, internal processes. | ||
| + | |||
| + | Individual trail marking. Navigate based on the aggregate distribution. | ||
| + | |||
| + | Do something that is more behavioral than rational, People react to whatever is not part of the daily business or of the routine, they REact to something and give a response. Ex. put an audio prompt on Slack, people can look at it or simply react to it. No need to prompt people to understand that that audio artifact is part of a research, just limit it to the reaction of people. | ||
| + | |||
| + | Tim: people need to connect in new ways, not just based on values, rationality, ideology. Emphasize on interactions, behavior, which has been lost in the age of modernity. | ||
| + | |||
=Some technical aspects= | =Some technical aspects= | ||
Revision as of 02:12, 8 October 2025
- Stigmergy is a mechanism of indirect coordination, through the environment, between agents or actions.[1] The principle is that the trace left in the environment by an individual action stimulates the performance of a succeeding action by the same or different agent. Agents that respond to traces in the environment receive positive fitness benefits, reinforcing the likelihood of these behaviors becoming fixed within a population over time. Stigmergy is a form of self-organization. It produces complex, seemingly intelligent structures, without need for any planning, control, or even direct communication between the agents. As such it supports efficient collaboration between extremely simple agents, who may lack memory or individual awareness of each other.
Applied to humans
- On the Internet there are many collective projects where users interact only by modifying local parts of their shared virtual environment. Wikipedia is an example of this. The massive structure of information available in a wiki,[19] or an open source software project such as the FreeBSD kernel could be compared to a termite nest; one initial user leaves a seed of an idea (a mudball) which attracts other users who then build upon and modify this initial concept, eventually constructing an elaborate structure of connected thoughts. In addition the concept of stigmergy has also been used to describe how cooperative work such as building design may be integrated. Designing a large contemporary building involves a large and diverse network of actors (e.g. architects, building engineers, static engineers, building services engineers). Their distributed activities may be partly integrated through practices of stigmergy.
Analysis of human social movements
- "The rise of open source software in the 21st century has disrupted the business models of some proprietary software providers, and open content projects like Wikipedia have threatened the business models of companies like Britannica. Researchers have studied collaborative open source projects, arguing they provide insights into the emergence of large-scale peer production and the growth of gift economy."
Stigmergic society
- "Heather Marsh, associated with the Occupy Movement, Wikileaks, and Anonymous, has proposed a new social system where competition as a driving force would be replaced with a more collaborative society. This proposed society would not use representative democracy but new forms of idea and action based governance and collaborative methods including stigmergy. "With stigmergy, an initial idea is freely given, and the project is driven by the idea, not by a personality or group of personalities. No individual needs permission (competitive) or consensus (cooperative) to propose an idea or initiate a project." Some at the Hong Kong Umbrella Movement in 2014 were quoted recommending stigmergy as a way forward."
Stigmergic organizations
As organisations, OVNs are self-organized through stigmergy, which is the primary mode of coordination. In the context of Sensorica, the digital environment where agents or affiliates collaborate is designed to enhance stigmergic behavior. For example, a shared online document can represent a free form collaborative environment. Sensoricans have created templates of such documents that contain a list of digital pheromones, called signalization tools, that agents can use to signal to others agents what has been done, where things are and what can be done to get wherever they want to get. See example of template for collaboration on technical development.
Tibi identifies different levels of stigmergy.
At the first level we have passive stigmergy, which is the type of coordination that ants/insects use. Agents (an ant) need to stumble randomly open a pheromone in order to get instructions what to do next. Pheromones that are placed far away, removed from the immediate context of activity, are not available to a particular agent. The system of agents works well if there are a critical number of agents, so that the probability that other agents stumble on a key pheromone is relevantly high. Since every agent only has access to its immediate environment, there is no way for a local agent to prioritize other than based on local ques/signals.
At the second level, we have active signaling stigmergy, which means that agents can get notifications from ques that are deposited in other contexts, based on some criteria. This second level was implemented in Sensorica's web2 version of the NRP-CAS. For example, if an affiliate has performed tasks in a context of work (i.e. a project or a venture), whenever there's another task of the same type that is created by another affiliate, the first affiliate can get an email notification. This allows a group of affiliates in a smaller size network perform well, by increasing the probability of an action being taken (someone will take a new task) in a different context of work, relaxing the requirement of the critical mass (high number of agents). This is a dumb as a matchmaking system. The idea is to reduce the distance / proximity requirement.
At the third level, we have what Tibi calls proactive stigmergy. We can also understand this as part of the feedback system. One way to implement this is to make the collaborative digital environment smarter, using machine learning or AI agents. In other words, AI agents can scan activity and flows within the environment and detect priorities in some contexts of work and send signals / ques / notifications to specific agents who can, directly or indirectly intervene. At this level we increase the general awareness about important issues throughout the network and further reduce the requirement for a critical number, allowing even smaller networks to operate well.
Stigmergy and Governance
Notes from meeting with the Metagovernance project and TaoDao. In the room: Tibi, Cent (Metagovernance), Daniel (Active inference lab), Ronen (metagovernance), Vini (TaoDAO), Tim.
Exploring governance and stigmergy, how these two conjugate together. Commons sense makers, stigmergy markings, open source attention.
In a social dynamic heavily driven by stigmergy rules are important but group decisions are disappearing.
We are exploring the interplay between governance and stigmergy,
Can we create a stigmergy / governance workshop? Vini says that people need to be able to reframe their attention and interest, in order to be able to sense their environment, the ecosystem. One requires an enhanced perception of the environment.
Set up a game field, see who picks it up, and those who come to the workshop, will be already selected by their awareness. Give time to incremental amounts of people. It can be an asynchronous experience. Make it easy to engage. Necessary to move the attention towards the ecosystem.
Stigmergy is like governance focused on attention. Governing our collective attention.
For stigmergy to work there is the network and the environment, and pressure from the environment that drives self-organization and optimization through stigmergy. There is a pressure for selection on the group, internal processes.
Individual trail marking. Navigate based on the aggregate distribution.
Do something that is more behavioral than rational, People react to whatever is not part of the daily business or of the routine, they REact to something and give a response. Ex. put an audio prompt on Slack, people can look at it or simply react to it. No need to prompt people to understand that that audio artifact is part of a research, just limit it to the reaction of people.
Tim: people need to connect in new ways, not just based on values, rationality, ideology. Emphasize on interactions, behavior, which has been lost in the age of modernity.
Some technical aspects
Quantitative vs qualitative stigmergy
Develop... see paper for definition
Anticipatory stigmergy
Stigmergy builds on information on the past. Anticipatory stigmergy is about agents anticipation of near future action. open paper