Frictionless Standards

Lightweight yet comprehensive data specifications.

The Frictionless Data project is built on top of the Frictionless Standards, which are a set of specifications created to standardize different aspects of working with data. For example, you can use the Standards to describe a collection of data files or to share information about data types.

TIP

This document is an overview of the Frictionless Standards - for more in-depth information please visit the specification site (opens new window) or click on one of the standards below and you will be redirected to a corresponding specification.

Standards Toolkit

At the core of Frictionless is a set of patterns for describing data including Data Package (for datasets), Data Resource (for files), Table Schema (for tables), and also domain-specific extensions.

Data Package

Describe a collection of data files in a single container called a package.

Data Resource

Describe and package a single data resource such as a table or file.

Table Schema

Declare a schema for tabular data that is expressible in JSON.

Fiscal Data Package

Describe and publish tabular fiscal data.

Data Package Views

Describe views of data by leveraging existing specifications like Vega.

CSV Dialect

Describe the various dialects of CSV files in a language agnostic manner.

Which standard is right for me?

To help you pick a standard to use, we’ve categorized them according to how many files you are working with.

Collection of Files

If you have more than one file:

• Data Package: Use a Data Package (opens new window) for describing datasets of any file format. Data Package is a basic container format for describing a collection of data in a single “package”. It provides a basis for convenient delivery, installation and management of datasets.
• Fiscal Data Package: For fiscal data, use a Fiscal Data Package (opens new window). This lightweight and user-oriented format is for publishing and consuming fiscal data. It concerns with how fiscal data should be packaged and providing means for publishers to best convey the meaning of the data - so it can be optimally used by consumers.

Individual File

If you need to describe an individual file:

• Data Resource: Use Data Resource (opens new window) for describing individual files. Data Resource is a format to describe and package a single data resource of any file format, such as an individual table or file. It can also be extended for specific use cases.
• Tabular Data Resource: For tabular data, use the Data Resource extension called Tabular Data Resource (opens new window). Tabular Data Resource describes a single tabular data resource such as a CSV file. It includes support for metadata and schemas to describe the data content and structure.
• Table Schema: To describe only the schema of a tabular data file, use Table Schema (opens new window). Table Schema is a format to declare a schema for tabular data. The schema is designed to be expressible in JSON. You can have a schema as independent metadata or use it with a Tabular Data Resource.
• CSV Dialect: To specify the CSV dialect within a schema, use CSV Dialect (opens new window). This defines a format to describe the various dialects of CSV files in a language agnostic manner. This is important because CSV files might be published in different forms, making it harder to read the data without errors. CSV Dialect can be used with a Tabular Data Resource to provide additional information.