Content Discussion History

An Open Data Program is as Successful as its Publication Process

Today, there’s nearly unanimous agreement on the benefits of open data, including transparency, accountability, and innovation, as well as greater citizen participation and access. And yet, there are some governments that haven’t seen the great gains that open data brings, even after releasing it to the public.

One main reason why your state or city might be missing out on open data success is because of your current publication process. If the process is ineffective, the outcome will not meet expectations. Here are three pieces of advice to help you plan and follow an open data publication process:

1. Develop a checklist of requirements to ensure that only appropriate data is published.

Having a checklist is the best way to make sure your data sets don’t include items that shouldn’t be disclosed. Some of these include political affiliation, religious beliefs, trade union membership, physical or mental health, etc.

  1. Specify which datasets that shouldn’t be released at all or, which may need further prep before their release (security, sensitivity, privacy, or data quality concerns).
  2. Classify datasets as restricted if they contain protected or private information that could create security vulnerabilities.
  3. Ensure datasets pass at least both privacy and security tests.
  4. Identify datasets collected or created by your city and outline the source, quality and revisions necessary before releasing to the public.
  5. Prioritize datasets that support your city’s strategic priorities.

2. Determine data access levels

  1. Remove barriers to data that can be released.
  2. Prepare the data by including a description of the variables and observations in the datasets.
  3. Ensure internal users are trained to manage the data workflow from collection, to publication, to public engagement.  
  4. Ensure all the datasets intended for release are machine readable, downloadable and in a usable format. (Tabular (spreadsheet) data should be published as comma-separated values (.csv) in clean sheet format)
  5. Provide a search function for people to find specific data.

3. Specify publication timeline and frequency of data updates

Except for budget data and time specific data, datasets will need to be regularly updated to remain useful. Develop a plan and publication schedule for adding new data and updating existing datasets. Remember, a vital part of the publication process is to announce new datasets and updates. Use multiple channels including mailing lists, blogs and newsletters to announce newly published datasets. Plan how to collect feedback from the public and track public engagement with your data.

Data publication is an “all hands-on deck” process and requires a comprehensive approach. Involve key staff and experts to determine compliance requirements. Data requires involvement with agency/department systems. When possible, involve stakeholders early in the publication process. Finally, specify free access policies and all appropriate data license where necessary. Remember, the most successful open data relies on preparation.



add comment

Your comment will be revised by the site if needed.