Highlights

• The co-production coherence can be verified by analyzing clues of agents language, its beliefs and its actions.

• Co-production of reference concepts and symbols promote acceptance of community rules.

• Collective beliefs shape colletive knowledge.

• colletive knowledge has the potential of developed

Where and Why you can use this meta-model

• A guideline for build a systemic governance model, with evidences based in a symbolic approach.

• A guideline to build a system indicators that use evidences from culture and power of agents.

• A coherence index to analyse consortions, suplay chains,

Agents and its befiefs

In this context, understand mechanism of beliefs is fundamental to identified indicators in order to measure the coherence of a production chain and block Eclipse Set. One way to identify this mechanism is mapping and stablish concepts, and values into symbolic representation models. In socio-technical system, symbolic representation models support business management due to efficiency, expressiveness and interpretability of represent processes and concepts.

An agent is a person or a computational system, autonomously capable acting in different contexts. Agents have this ability because their dynamics mental states, created from beliefs, desires and intentions [8–10]. Belief is an intentional state of mind, which represents the knowledge about the world [11]. Human agents are referential for computational agents cognitive modeling. In artificial agents, beliefs are considered a symbolic state representation of a possible world, which can be described in terms of variables, database, computational ontologies or predicate logic.

For agents (human or digital system), beliefs represent actions experience into a dynamic world, into some specific situations. Those experiences are record in agents minds through the symbols [8–12]. Agents can transform what they believe, their world view, from the symbols they internalize, and consequently, agents can change their actions on the world. Therefore, beliefs play a fundamental role in knowledge creation, but it is severely ignored in production and knowledge management. As Bourdieu [13] describes, beliefs are formed where dialogs are formed, in cultural interaction and in social media. Beliefs can be both, a product and an initial source of collective knowledge. Human agents needs to know if they share the same beliefs and symbolisms of their possible worlds or their perspective of what is possible and morally acceptable with other agents [8,14,15].

Symbols

Selecting the right types of symbolic representations for coproduction tasks, allows structural alignment by similarity. Thus facilitating the learning of a system. In terms of evaluations, symbolic representations are more expressive and efficient than a purely statistical model [5]. The integration of symbolic representation into contracts for governance can benefit from semantic technologies (ontologies) as well as security technologies such as Blockchain [6]. However, studies are still needed to establish socio-technical environment capable to make integration of the negotiation of meanings with these technologies [7].

In order to minimize the Eclipse Set, and make collective knowledge accessible to all agents (humans, machines or both), this article present CoheNet Metamodel, which aim to establish a common ground to a socio-technical community system. Common ground present in this research explores symbolic representations from the pragmatic level of production chains, intensive in knowledge and technologies (socio-technical complex system). These representations aggregate metadata structures to accommodate technologies such as Blockchain and ontologies, aiming to produce value with human and artificial agents. Using the logic of the practical rationality of an artificial or human agent, this study presents metrics of coherence between reference parameters and situational solutions, though a metamodel based in functional systemic linguistics. CoheNet metamodel is intended to support virtual heterogeneous agents interaction, with characteristics such as, diversity of technologies, beliefs, values, and languages, who come together to co-produce common goods, and have to deal with problems that conceal the productive knowledge (Eclipse Set) [3].

Design Science Research Methodology

Metamodel proposed was developed from a Design Science Research (DSR) methodology,ilustrated in Figure 1

Figure 1 – Design Science Research Cycles from Schneider (3) apud (34).

Using methods of integrative revision of literature, systematic reviews and exploratory research, Schneider [3] trailed an approach from the deductive method, which part of the following premises to form a conclusion:

• (premise 1): people unite to co-produce common goods because they believe in symbolic entities consistent with their beliefs [2,8,14,15, 20,21, 26, 29];

• (premise 2): Pragmatic Web supports the negotiation of common symbolic entities by presenting methods and techniques for the co-production of meanings and values between human and artificial agents [1,6,22–24];

• (conclusion): symbols described from Pragmatic Web methods and techniques unite agents in a common good co-production, in sociotechnical contexts.

Pragmatic Web guidelines aims to support coproduction actions, in a phase prior to creation of artifacts, which provide understanding of concepts for systems, humans or both. For artificial agents, those artifacts can be algorithms, taxonomies, techniques, metadata architectures, or each thing can be transformed into inputs for databases, computational ontologies, predicate logic or smart contracts (with blockchain). Furthermore, for human agents, those artifacts can be guides, manuals, contracts and agreements [22,31–33].