This site is the home of simple protocols and formats for working with open data. Our mission is both to make it easier to develop tools and services for working with data, and, to ensure greater interoperability between new and existing tools and services.
Work topics include:
- Data packages, publication and installation
- Simple formats for facilitating data exchange over the web
- Web-oriented data APIs including those for accessing and querying data
- An overview of sharing and syncing data changes
- Webhooks and webservices for data transformation
- Data catalog interoperability - see the Data Catalog Interoperability Protocol
Table of Contents
- Data Packages
- Tabular Data Package
- Data Package Identifiers
- JSON Table Schema
- CSV Dialect Description Format (CSVDDF)
- Coopy highlighter diff format for tables
- Units for Data
- CouchDB Replication Protocol
- Linear TSV
- NDJSON - Newline delimited JSON
Research and Patterns
Background and Context
The civic and open data community is often hampered by a lack of simple standards for interaction between services and tools.
We believe there is a clear need for the kinds of lightweight but useful protocols and formats for doing things like:
- Sharing information between data catalogs
- Packaging data for publication and installation
- Sharing and syncing data changes over HTTP
- Querying data and databases over the web
- Creating web services for data transformation and reconciliation
DataProtocols isn’t a formal standardization process like W3C, ISO or OASIS, but rather is aiming to be a more informal community where we can hammer out RFCs for data - rough consensus, running code and integrated data (and building wherever possible on what already exists).
We believe consensus needs to come from a community not just a single vendor or organization and the user and developer community is now developed enough to take this on and make it happen.
Moreover, technology and especially web technology has reached the point where this is both feasible and needed - be that in terms of standards, use of HTTP and REST, browser maturity, or, in terms of ad-hoc development in tools that is ripe for “standardization”.
Data Protocols was started in Autumn 2011 and arose out of discussions with, among others, Rufus Pollock and Friedrich Lindenberg of the Open Knowledge Foundation, Francis Irving and Aidan McGuire of ScraperWiki, Max Ogden (then a Code for America fellow), Chris Taggart of OpenCorporates, Richard Cyganiak of DERI and members of the W3C GLD Working Group and has subsequently benefitted from input from numerous other individuals.
Participate and Contribute
This effort is community-run and contributions, comments and corrections are warmly welcomed. Most work proceeds in an RFC-style manner with discussion on the mailing list.