Last month, we had the first call of the Frictionless Data Specifications Working Group, starting a new chapter in the project. The call covered the status of the specifications to date, current adoption, upcoming technical pilots and partnerships, and how work will be organized going forward. In this post, I will lay out the purpose for this initiative, who is participating, and how you can get involved.
Frictionless Data is a project encompassing a set of tooling and specifications to ease the transport and reuse of data. The specifications have grown out of a long engagement with issues around data interoperability, publication workflows, and analysis. For most of the history of this project, the specifications were curated by Rufus Pollock as one of several “Data Protocols” with input and assistance from individuals from Open Knowledge International and other organizations. As a result, the specifications have steadily gained traction across various projects and software developed by, among others, the Open Data Institute (ODI), Tesera Systems, Inc., Dataship, and Open Power System Data.
Members of the working group currently include:
- Paul Walsh (Open Knowledge International)
- Rufus Pollock (Open Knowledge International)
- Dan Fowler (Open Knowledge International)
- Dominik Moritz (University of Washington)
- Steven De Costa (Link Digital)
- James McKinney (Open North)
- Karissa McKelvey (Dat Data
- Spencer Cox (Tesera Systems, Inc.)
Work will happen continue to happen asynchronously, in the open, without excessive rules around voting. Rather, we will listen to feedback and act in favor of consensus (without requiring it). Rufus Pollock, having led this work for many years with a strong focus on keeping it simple, will remain the curator; decisions of what stays or goes from the specs will rest with him. Having more eyes the specs with a variety of different perspectives will allow us to solidify and remove ambiguous statements, eliminate unnecessary repetition or logical errors, and, hopefully, achieve a minimal 1.0 by end of 2016. Beyond the core Data Package specifications, open topics might include defining further custom “profiles” (e.g. Fiscal Data Package), as well as potential extensions, including specifications for visualizations, statistics, and quality metrics for data.
Are you currently using or considering using the Frictionless Data specifications for your data or application? If so, please let us know!
Work is managed via an issue tracker on GitHub, which is the best way to raise specific questions. If you would like to specifically flag an issue for the Working Group, mention @frictionlessdata/specs-working-group in the comment. For general commentary on any aspect of Frictionless Data, you can leave a comment on the forum.
- Current Specifications: http://specs.frictionlessdata.io/
- JSON Schema (for validation): https://github.com/frictionlessdata/schemas
- Specs Issue Tracker: https://github.com/frictionlessdata/specs/issues
- Current Milestone: https://github.com/frictionlessdata/specs/milestone/1
- Forum: https://discuss.okfn.org/c/frictionless-data
Thanks to Paul Walsh who provided the motivating text that served as the basis for this post and Jo Barratt who did much of organizing necessary to make it happen.