When and how to publish open data

In keeping with the right of access outlined in Ontario’s Municipal Freedom of Information and Protection of Privacy Act, divisions should make data open whenever there is demonstrated public interest in the information, when publishing that information could create positive impacts for Torontonians, and when there are no privacy concerns with releasing that data.  

The following documents guide how divisions should publish open data:    

• The City’s Open Data Policy, which asks divisions to “identify existing and potential datasets for release as part of the Open Data Program and work with the Open Data Team on the planning and development of new datasets, review of existing ones, publication of datasets, and archiving of superseded datasets.”   
• Council directive 2019 GL8.22, which asks that data “embedded in documents, reports, or any digital artifacts available publicly on the City of Toronto’s digital infrastructure” be made available as open data;   
• Council directive 2021 EX22.13, which mandates that City-owned or managed data used in staff reports to committees or Council be considered for publication as open data.  

Process overview

No two datasets are exactly alike, and the steps, effort and time required to publish an open dataset can vary.

In cases where divisions have high quality data and where any concerns about privacy have been addressed, a new dataset can be published in about two weeks.  More complex cases can take longer, but the Open Data Team can provide support and resources to help.  

Publishing open data typically involves the following steps:   

1. Identification: Finding and prioritizing datasets for publication in line with relevant Council directives (GL8.22; EX22.13), the City’s Open Data Policy and relevant legislation (City of Toronto Act, Municipal Freedom of Information and Protection of Privacy Act, Personal Health Information Protection Act, etc.).  
    
   This process involves reviewing your division’s data assets, identifying potential open datasets, and prioritizing them for publication based on public demand, potential for positive impact and suitability.  

2. Development: Deciding on the structure and composition of datasets and preparing them for publication according to the Open Data Guidelines.   
    
   This involves formatting the data, crafting the appropriate metadata, removing or de-identifying sensitive and personal information, and crafting a narrative to help users understand and make use of the data.    

3. Publication: Loading data onto the Open Data Portal from its source system.   
    
   This involves creating a pipeline to get the data from its source system to the Open Data Portal. This may be done manually by staff, or by using an automated process.   

4. Maintenance: Ensuring the dataset remains accurate and useful over time.   
    
   This can involve updating the data according to its stated schedule, addressing any concerns about accuracy or quality, improving or automating the data pipeline, or responding to public inquiries about the data.  

5. Retirement: Clearly communicating when a dataset is no longer active or being maintained. 
    
   If an open dataset is no longer being updated or has been replaced by new data with the same or similar information, it should be marked as retired. Retired datasets remain accessible on the portal but are clearly marked so users understand the data is no longer current.  
    
   Open datasets may also be deleted from the portal in specific circumstances.  

Working with the Open Data Team

The Open Data Team assists divisions at every step of the publishing process. Whether you’re learning about open data for the first time, curious about how to protect sensitive or personal information in your data before opening it, or need to make changes to an existing open dataset, we’re here to help!