Note: This content is not yet available through gitbooks. GovEx team members can make direct edits to this file. Other partners should use the following approaches:
- propose direct edits to this file (will automatically fork and create pull request)
- fork this repository, make edits, and send a pull request
- create issues in this repository which the GovEx team will resolve.
Github accounts are required.
This document contains a set of sample requirements to help governments evaluate, develop (or procure), deploy, and launch an open data web site (portal).
The following terms are used throughout the requirements to indicate importance.
- Must denotes a requirement that is critical to the successful implementation of the open data portal. If this requirement is not met, it will significantly impair the use or management of the site.
- May denotes a requirement that is optional, but highly desirable for success.
The following roles are used throughout this document to approximate to whom each requirement is relevant. They are not intended to reflect a permission system, which might be implemented by a technology solution.
- A customer is anyone who visits the portal to find, and access, and use data.
- Typically, a publisher is a government employee who is responsible for publishing or updating data and other content on the portal.
- Typically, an administrator is a government employee who manages the technical and configurableconfiguration options of the portal, such as visual themes, metadata requirements, etc.
Generally, a publisher can do anything a customer can do, and an administrator can do anything a publisher can do.
A data catalog is a listing of available data, with each catalog entry corresponding to one or more data resources. It usually includes high-level information for each dataset, such as Title, Description, Release Date, Category(-ies), and Keyword(s).
- must be able to search for datasets by one or more terms contained in dataset metadata
- must be able to browse or explore data by category
- must be able to download the catalog in a machine-readable format compatible with the common core metadata schema from a well-known address (typically /data.json)
- must be able to view a summary page for each dataset which details data resources, metadata, and other relevant documentation
- may be able to search for datasets by terms contained in the data
- must be able to add, edit, or remove catalog entries
- may be able to mark catalog entries as private so they are not visible to the public
- must be able to create, edit, and retire metadata categories
- must be able to configure default sorting for data catalog
- must be able to configure look-and-feel of data catalog
- must be able to download stored data in bulk
- may be able to access stored data through a REST API
- must be able to create new data resources
- must be able to replace or update existing data resources
- must be able to manage permissions for other administrators and publishers
- must be able to preview data resources using an through their internet browser (for example, through a rows-and-columns grid or a map)
- may be able to visualize data resources in a graphical way, including maps, graphs, charts, etc.
- may be able to save visualizations with configuration settings for future viewing
- may be able to create data visualizations and make them accessible through the data catalog or data resource page(s).
- may be able to control whether customer-created data presentations are visible to other customers
- must be able to provide comments and feedback on individual data resources
- must be able to nominate data resources for public release
- may be able to provide comments and feedback on the data catalog
- must be able to view customer-nominated data resources
- may be able to update the status of customer-nominated data resources
- may be able to respond to customer-nominated data resources
- may be able to view and respond to customer comments and feedback
- may be able to review and approve visibility of customer comments and feedback
- must track and analyze customer feedback
*may be able to register for an account with an email address *may be able to create, edit, or remove profile information which may be accessible to other customers
- may be able to grant specific, registered customers permission to access specific data resources that are owned by that publisher
- must be able to manage permissions for other administrators and publishers
- must be able to disable customer accounts (if customer accounts are a feature of the portal)
- The following requirements are general technical requirements which either don’t require a specific role or apply to all roles.
- Data resources must be available for download in the original format which they were published on the portal.
- API calls which are invalid or fail must return the appropriate HTTP response status code (404, 500, etc) and status message.
- The portal must be accessible through a city-provided Internet domain name (e.g. “data.somecity.gov”).
- Transport Layer Security (TLS) must be used for account registration, logging in, and all actions performed by publishers and administrators.