This is a template repository for the Reusable Building Blocks (RBB) Initiative. See https://agentblocks.org for the official website.
Additional discussion and context can be found in our paper in Environmental Modelling & Software: "Towards reusable building blocks for agent-based modelling and theory development".
RBBs are submodels that describe processes relevant for a broad range of ABMs in a certain application domain, for example plant competition in vegetation models or reinforcement learning in a behavioral model.
This is still a work in progress, and we are grateful for any comment or suggestions about the following two-tier structure for RBBs. Tier 1 requirements are intended to set an initial low bar for submission and sharing so that good ideas can be shared and improved upon. Tier 2 requirements establish a rigorous, tested baseline for RBBs that can be peer reviewed and published in a trusted digital repository with a DOI similar to a journal submission.
The RBB's name should be self-explanatory and short.
A narrative description that addresses the following:
- What mechanism or process does this RBB represent?
- What kinds of environments, systems, and research questions has this RBB been used for, or could it be relevant for? Include durable references (e.g., DOIs, permanent URLs, or other permanent identifiers) to past work.
A detailed narrative description of this RBB describing, i.e., its entities, processes, parameters, and scales according to the rationale of the "Overview" part of the ODD Protocol.
A visual representation of the RBB should also be strongly considered to be included, e.g., a flowchart and/or pseudocode that demonstrates the essential mechanism(s) of the RBB.
An implementation of the RBB in a specific programming language and/or software framework provided in a readily executable format (e.g., source code that can be easily compiled into an executable)
- well commented
- clear metadata that describes what programming language, version, operating system, and framework(s) used
- the executable demonstration program that shows how the RBB works and allows users to interact with its inputs and parameters
A comprehensive list of all inputs and outputs of the RBB, i.e., all variables it needs as the basis for its calculations, and all variables it changes as a result.
Durable references (i.e., permanent identifiers or URLs) to other related RBBs.
Tier 2 RBBs require documentation of the RBB in the spirit of the ODD protocol for describing ABMs. After reading Tier 2 RBB documentation a human should be capable of re-implementing the RBB themselves and completely understand what the RBB is doing and how.
Tier 2 RBBs should meet all Tier 1 requirements with the following additions as well as more detailed descriptions of a few of the Tier 1 fields.
relevant keywords or tags for this RBB
- authors
- version
- license
- programming language and version
- software, system, and data dependencies
- repository URL (NOTE: needs clarification, is this the development repository like the URL of this GitHub repository, or a TRUSTed digital repository for archival, like https://doi.org/10.5281/zenodo.7241586?)
- peer reviewed (yes/no)
- date published
- date last updated
Create a citation for this RBB according to the Software Citation Principles.
This can be easily adopted by editing the CITATION.cff file in this repository before publishing this RBB in Zenodo.
Then, edit this section and indicate how you'd like for this RBB to be cited, e.g.,
Druskat, S., Spaaks, J. H., Chue Hong, N., Haines, R., Baker, J., Bliven, S., Willighagen, E., Pérez-Suárez, D., & Konovalov, A. (2021). Citation File Format (Version 1.2.0) [Computer software]. https://doi.org/10.5281/zenodo.5171937
Be sure to update the citation manually if you go this route and the citation changes.
Describe the theoretical concepts the RBB is building on, e.g., a specific standard description of a certain behavior such as symmetric competition or the informed user. This helps to classify the building block and to develop links to other similar RBBs.
RBBs are typically relatively short procedures that describe a certain process or interaction. They typically affect a certain entity, e.g. an agent or a spatial unit, and they are typically run by a certain calling agent, or agents. For example, a plant uses water, or competes with other plants, a buyer buys somthings, or a farmer is affecting the way how other farmers use their land. Note that the calling and affecting agents can be the same.
- What entity, or entities, are running the submodel? What state variables does the entity need to have to run this RBB?
- Which variables describe the entities (normally derived from state variables)
- What global variables (e.g., parameters characterizing the environment) or data structures (e.g., a gridded spatial environment) does the use of the RBB require?
- Does the RBB directly affect global variables or data structures?
- Which parameters does the RBB use? Preferably a table including parameter name, meaning, default value, range, and unit (if applicable)
Patterns and data to determine global variables & parameters and/or to claim that the model is realistic enough for its purpose
- Which of the variables (or parameters) have an empirical meaning and can, in principle, be determined directly?
- Which variables can be only determined via calibration?
- Which data or patterns can be used for calibration?
- Which data sets already exist? (include durable references)
- What specific inputs does the RBB require from an external, calling entity and in what units (e.g., CSDMS Standard Names and UDUnits)?
- What specific outputs (also specifying units and as much semantic information as possible) does it produce and how should these outputs be interpreted to update the state variables of the calling entity?
- On which spatio-temporal scales does the RBB work, i.e. what are the resolution and extent of the spatial and temporal scale?
How, in detail, does the RBB work? This should be a narrative description that describes the code implementing the RBB and can include equations and pseudo-code which is particularly important if the RBB involves several processes.
- include a flowchart if appropriate
Provided in a format that is readable by compilers/development environments, well commented, written for which programming language, operation system, version etc., if possible, provided for different programming languages
An executable deployment that includes a simplified environment in which the RBB can be run.
Results obtained with the example implementation providing insights into how, under different settings, the RBB performs, including extreme scenarios.
What is the history of the RBB? Is it entirely new or based on earlier submodels, or an implementation of an existing submodel?
Has the RBB, or its predecessors, been used in the literature?
Include a reference list of publications where the RBB was successfully used.
For more complex RBBs, a detailed user's guide or manual and a tutorial walk through can be very helpful to onboard new users.
As needed.