Opening this issue to track thoughts on open source data schemas/standards/formats.

General Questions

• How are people in the world implementing open-source data standards/schemas and moving data around?
• Do special considerations exist for (non-PII, non-confidential) federal spending govt. data standards that distinguish this space from data standard implementations in other industries?
• How important is it to build validations/metadata into a schema versus having a separate validation process and metadata definitions?
• Related: what are the pros and cons of building a data input/validation process tied to a particular schema syntax vs. de-coupling syntax from validations/metadata?

Questions for specific data standard/schema options:

• How robust is its open-source ecosystem? Has the open source community developed tooling around it? If so, how actively are the tools updated/maintained, and how large is the contributor base? Will users incur licensing fees or get locked into a vendor?
• How widely adopted is the data standard/schema? Is it limited to a particular domain?
• How easy is it for data submitters/producers (both biz and tech) to use?
• How easy is it for data consumers (both biz and tech) to use?
• What resources exist for non-coders to understand the data standard/schema? Is that even relevant if good, robust metadata exists (see above)?
• What's the process for updating and tracking changes to the schema/standard?

These are schemas/standards/formats we know about. What are others?

• XML/XBRL
• JSON/JSON-LD/JSON Schema
• Protocol Buffers
• Avro
• Thrift
• Message Pack
• ORC --> likely not relevant but here for completeness
• Parquet --> likely not relevant but here for completeness