Lessons Learned
Although testing and deploying new and interesting technical tools for the launch of the portal was an experience to itself, our best lessons learned came from other City staff we worked with. Because the Open Data portal was an entirely new effort for the City, we had to be open to change and refine our goals, expectations and workflow as we went. We aim to increase data literacy, manage data as an asset, ensure data is described and catalogued according to established standards, support increased use of data in decision-making, and encourage innovation -- and this means working closely with people to demonstrate the possibilities of Open Data.
City employees are enthusiastic, but also careful, about releasing data
City staff recognize the benefits of opening up data, but they want to make sure to provide quality data that the public can understand and analyze correctly. This concern is one that we take seriously, and we began addressing it right from the beginning. The portal offers a variety of ways to add extra information and context on the dataset page where people download the data. We also offered our colleagues the option to create data stories that provide context. Staff want to be involved every step of the way to make sure the City provides the right data, in a correct format, with plenty of accompanying information to make sure people don’t misuse or misinterpret the data.
City staff will be some of the biggest users of open data
Many, many of our internal partners throughout the City have approached us with ideas for analyzing and displaying their data beyond just offering a download to the public. The Data Stories published on the portal at launch were just the tip of the iceberg of ideas. Fellow City employees also were a significant portion of people who voted on the inventory at datasd.org.
City staff need help using technology to improve their data collection and management processes
City employees who have ideas about how they can improve their data management naturally have approached the Chief Data Officer for help and guidance. Sometimes, people just need a little bit of training to increase their technical skills. Other times, sections and divisions and departments need significant help using technology to become more efficient and solve problems.
One success story we can already tell is the online Public Records Act Requests portal at sandiego.nextrequest.com. The CDO and Performance & Analytics Director Almis Udrys participated in the working group that tested solutions for making the PRA response process more efficient in the City, and Performance & Analytics helped the Department of IT procure this product. The PRA tool now boasts nearly 2,000 successful responses to records requests. We have started brainstorming other similar opportunities to improve data collection and management with the Fire-Rescue Department, the Police Department, traffic management, economic development, the Climate Action Plan oversight team, and the traffic counts program.
Releasing data is not simply getting the login to a database or access to a spreadsheet on a shared drive
The City has always wanted to release well-described, high-quality data, but our team set a standard that we would not include anything on our portal Performance & Analytics could not answer questions about. We didn’t just leave it up to information coordinators in departments to describe the data and check off the box when they submitted a data dictionary; we attempted to learn about all aspects of the data, even if that meant refreshing our knowledge of science or gaining in-depth knowledge of campaign finance laws.
One rigid data release process is not effective if you want a catalog of high-quality data
Departments and City staff determine their own methods for collecting, storing and updating data according to what works for them. Additionally, data is managed by people with varying technological resources and skills, and some data is under the management of a contractor or vendor. Imposing a rigid, and possibly disruptive, data release requirement will not get buy-in from other City employees and will limit our ability to accomplish the larger goals for our program. We had to be flexible enough to customize the process to each dataset, and be willing to make a few compromises as we continue to work with fellow staff.