How do we make our data
more accessible?

Key concerns

• Encouraging more use cases and public engagement
• Sustainability/ease of maintanence
• User-friendly interface

Pros and cons

A data portal does not preclude creation of dashboards!

What is data portal?

“A data portal is a site for storing a list of datasets with added features to facilitate accessing, viewing, and searching for any particular datasets.”

• Examples:
  • Data.gov by the US government
  • Chicago Data Portal by City of Chicago

Why data portal?

• Proven utility
  • Many use cases (e.g. Data.gov, Chicago Data Portal)
• Great for storing/sharing multiple datasets
  • Suitable for the R&A unit’s “data clearinghouse” mission
  • Can have various formats of datasets (even dashboards)
• Ease of use and management
  • User interface
  • Access control

What is CKAN?

• The Comprehensive Knowledge Archive Network:

“CKAN is an open-source DMS (data management system) for powering data hubs and data portals. CKAN makes it easy to publish, share and use data.”

Use cases

• Data.gov by the US government
• European Data Portal by the European Union
• opendata.swiss by the Swiss government
• And more!

Data.gov main page

opendata.swiss main page

Why use CKAN?

• An actively managed/maintained open-source project
  • Free, flexible, and up-to-date technology
• A rich set of features for a fully-functionl data portal
  • Extensive documentation is available for various aspects/features
  • 200+ extensions exist to provide additional functionalities

Useful features

• Flexibility and customizability in portal structure and look
• Elaborate access control
  • Organizations and groups
  • Capable of supporting a sustainable workflow
• Built-in visualizations
  • A simple “dashboard” view for each entry is available via an extension
• API for programmatic access to data
  • Great for research and application building
  • An entirely separate front-end interface is possible!

General architecture

General architecture (cont.)

• Using data.illinois.gov CKAN server as a backend DMS/storage
• Using ShinyProxy to deploy and manage staff-made dashboards/applications
• Creating a front-end web application for customized user interface/experience
  • This is the “Data Portal” that people will see and utilize to access ICJIA datasets, publications, and dashboards

Portal structure

• Main page
  • Highlights, search bar, categories, some stats, etc.
• Data
  • Raw datasets (with metadata and simple visualizations)
• Publications
  • Articles
  • Dashboards (with metadata)
• Applications
  • Featuring user-made applications
• About
  • FQA, API documentation, Application guidelines, etc.

Access Control

Access control model is implemented via data.illinois.gov. Select staff members will gain adminisrator rights and work with DoIT to build access control model that reflects ICJIA’s organizational structure

• Users with permission
  • upload/edit an item for internal view only
• Mangers
  • upload/edit/delete an item for internal and public views
• Administrators
  • manage organizations and groups
  • grant permissions as needed

Contents

• Data entries
  • Raw datasets, dashboards, publications
  • Metadata: brief descriptions, data dictionary, etc.
  • Tags for grouping/classifying entires
  • Data explorer and visualization views
  • API for programmatic access/use of data
• Publications
  • Existing articles on the current website
  • New articles
    • interactive articles
  • Dashboards

Data entry; image from CKAN Documentation

“View” options; image from CKAN Documentation

“Data Explorer” view; image from CKAN Documentation

Built-in graph; image from CKAN Documentation

Embedding a webpage; image from CKAN Documentation

Article page example

User engagement

• Application programming interface (API)
  • Facilitating data access for research projects or building applications
• User-made applications
  • Featuring applications meeting our standard
  • Providing an incentive to use our data -> more use cases
• Open-source contributions
• Outreach effort
  • Raising awareness and offering trainings on our UI/API
  • Hosting a “data challenge”? (e.g. 2015 Divvy Data Challenge)

Applications page example

Internal timeline

Grant application timeline