Processing, publication, access, insight. The future of data is frictionless.

Specifications

Lightweight containerisation formats for data that provide a minimal yet powerful foundation for data publication, transport, and consumption.

Go to specifications

Software

Apps and integrations that make it easy to integrate Frictionless Data specifications into your data publication, access, and analysis workflows.

Go to software

Case Studies

Labs, libraries, governments, and companies are using Frictionless Data in their data workflows to reduce grunt work and move faster to insight.

Go to articles

Get involved.

Guides

Learn how to get started using and developing with Data Packages through our guides, tutorials and documentation.

Go to guides

Chat

We also have a chat room dedicated to technical discussion about Frictionless Data. Feel free to stop by and introduce yourself.

Go to chatroom

Forum

If you have a use case that you'd like to see supported, including integration with a particular tool or file format, leave a note.

Go to forum

Our Vision

Frictionless Data shortens the path from data to insight with a collection of specifications and software for the publication, transport, and consumption of data. At the heart of our approach is a deep understanding of the multi-faceted nature of data work, and an emphasis on platform-agnostic interoperability. From consumer spreadsheet software, through to cloud-based services for data analysis, the future of data is frictionless.

Informed by our work building and deploying CKAN and learning about various data publication workflows, we have learned that there is too much friction in working with data. The frictions we seek to remove---in getting, sharing, and validating data---stop people from truly benefiting from the wealth of data being opened up every day. This kills the cycle of find/improve/share that makes for a dynamic and productive data ecosystem.

We provide a simple wrapper and basic structure for transportation of data that significantly reduces the friction in data sharing and integration, supports automation and does this without imposing major changes on the underlying data being packaged. We focus on tabular data but any kind of data can be "packaged". Its lightweight and simple nature it is easy to adopt both for data publishers, data users and data tool makers.

We have been working on these and similar issues for nearly a decade, and we think that the time is right for frictionless data. Help us get there.

Data "Containerization"

We see our approach as analogous to standardization efforts in the transport of physical goods. Historically, loading goods on a cargo ship was slow, manual, and expensive. The solution to these issues came in the form of containerization, the development of several ISO specifications specifying the dimensions of containers used in global shipping. Containerization dramatically reduced the cost and time required for transporting goods by enabling the automation of several elements of the transport pipeline.

We currently consider transporting data between and among tools to be comparable to shipping physical goods in the pre-containerization era. Currently, before you can properly begin an analysis of your data or build a data-intensive app, you have to extract, clean, and prepare your data: procedures that are often slow, manual, and expensive. Radical improvements in data logistics---through specialisation and standardisation---can get us to world where we spend less time sorting through and cleaning data and more time creating useful insight.

Principles

1. Focused: Focus on one part of the data chain, one specific feature (e.g. packaging), and a few specific types of data (e.g. tabular).

2. Web-oriented: Build for the web using formats that work naturally with HTTP such as JSON, a common data exchange format for web APIs, and CSV, which is easily streamable.

3. Distributed: Design for a distributed ecosystem with no centralized, single point of failure or dependence.

4. Open: Make things that anyone can freely and openly use and reuse with a community that is open to everyone.

5. Built Around Existing Software: Integrate with existing software while also designing for direct use---for example, when a Tabular Data Package integration is unavailable, fall back to CSV.

6. Simple: Keep the formats and metadata simple and lightweight, and make things easy to learn and use by doing the least required.

bookdocsexternal fforumgithubgitterheartpackageplayrocket tools