Navigation
- Overview
- What are we trying to standardize?
- A living standard
- Standard data types (concepts)
- Version
- Organization
The Artificial Life (ALife) community is awash with disparate software systems, many of which produce data about the same underlying processes. This results in duplicated effort in developing analysis tools and visualizations. We believe we can fix this problem by developing a community consensus around digital standards for common data types (e.g., genomes, phylogenies, etc.). Our aim for these standards is for them to be useful to a wide range of artificial life systems, including artificial chemistries, abstract ecologies, evolutionary systems, and any other communities who express interest.
Eventually, this repository will hold a generally-accepted specification for these standards, but much discussion is sill required to get to that point. Currently, this repository contains proposed draft specifications for the standards and exists primarily as a home for further discussion to finalize them.
Our goal is to specify standard ways of describing and storing particular types of data. These standards do not describe methods for analyzing or working with data. By agreeing on standard ways of describing and storing data, it will become easier to share and make use of analysis tools and visualizations.
The ALife data standards should be a community effort, changing to conform to the needs of our community. Anyone is encouraged file issues on this repository to make suggestions for new standard data types, point out problems with existing standards, et cetera. Pull requests to make changes to standard documentation or to add new standards are also welcome.
Eventually, we should develop policies/templates for requesting/specifying standards. Suggestions for this are welcome!
Different systems and experiments need to represent different types of data (concepts). We aim to modularize those concepts into their own standards, ensuring that these standards as a whole are complimentary to each other. For instance, the entities in a phylogeny may be genotypes, but systems without phylogenies may also need genotypes. Instead of defining a standard for genomes in multiple places, we define it once and allow references to genomes to show up in other standards.
The standards that we have begun developing thus far are:
In the long run, there are more components of ALife systems that we hope ot have standards for, including:
- Meta-data (data that provides context for other data)
- Brains/controllers/information processing units
- Interaction networks (graphs indicating interactions between members of a population)
- Behavior (may or may not be practical, but worth discussing)
- Fitness landscape
- More?
The current version of the standards is 1.0-alpha.
- The most recent change to the phylogeny standard was in version 1.0-alpha
Standards-related materials all live inside repositories owned by the alife-data-standards organization. These materials are divided into two groups:
- The standards themselves, which live in this repository. Changes to the standards can be proposed in three different ways (to accommodate different levels of comfort with GitHub): submitting a pull request with the proposed changes to this repository, making an issue on this repository, or e-mailing the standards maintainers (listed below). Once a modification is proposed, discussion on it will take place via GitHub issues (which also supports contributions via e-mail). When consensus is reached, the change will be added to the official standards.
- A vetted collection of tools that work with standards-compliant data, which live in the tools repository. These tools include converters into and out of the standards format. If you have written a tool that you would like to add to this repository, you can request that it be added via pull request, issue, or an e-mail to the maintainers. The maintainers will then confirm that the tool works with the current version of the standards before accepting the addition. Updates to tools can be added in the same way. This repository also houses a list of third-party tools that work with data in standards format. These tools are not vetted or tested by the ALife Data Standards maintainers. You can request that a tool be added to this list with a pull request, GitHub issue or e-mail.
- Cliff Bohm (cliff.bohm@gmail.com)
- Alexander Lalejini (amlalejini@gmail.com)
- Emily Dolson (EmilyLDolson@gmail.com)
Find our code of conduct here.