This morning I met with councilor GT Bynum, Amy Brown, and Tom Golliver (thanks for the hookup Matt). I shared and summarized the policies of some cities powered by open source – esp. Vancouver on which I believe we should base Tulsa’s. I really like that Vancouver’s defines three distinct “open” policies:
Reduce costs when replacing or adding systems by considering open source software.
We know we can save money on software licenses, but we don’t want to lose money or time by forcing people to learn new software programs. So exploring and adopting open source software should be timed with plans to replace or add systems that already imply re-training users.
I was surprised and impressed at the list that Tom compiled of open-source software the city already uses, including WordPress and Umbraco. It just confirms that open source has “arrived” and is already established throughout government and enterprise IT. So a city policy is intentionally embracing an existing trend in city IT.
Adopt prevailing open standards for data, documents, maps, etc. to improve efficiency and coordination between local, state, and federal departments and agencies; create opportunities for Tulsa’s creative tech sector to improve transparency of, and access to, city information.
In my opinion this is the most important policy we can adopt quickly. Whether we use closed- or open-source software, all the city’s software should use open standards. It should be part of the software procurement process, and should be grounds for replacing existing software.
Many examples of open standards working well:
- Amy sent the meeting invite in iCalendar format which Exchange/Outlook, Google Calendar, and iCal all support; all of us were able to RSVP in our own calendar applications.
- TRIF was built using an RSS feed from City of Tulsa‘s Live Traffic Report.
- OttoZen was build using TRIF.
- Oklahoma Boundary Service was built using shapefiles from the OU Center for Spatial Analysis.
We’ve already worked on making non-standard data more accessible:
- We’re still trying to transform MTTA’s data into the GTFS standard to enable things like Open Trip Planner, Google Transit, and GTFSExchange.
- We wrote code to put public meeting agendas into an iCalendar feed.
- We’re scraping Tulsa Health Department’s Restaurant Inspection data to create a consumer-facing mobile app.
Because we had access to data in open standards, TRIF, OttoZen, and OBS were far easier to build than the latter projects.
3. Open Data
Publishing (more) data empowers citizens to create a more vibrant city and provides entrepreneurs with opportunities to create local businesses.
After open standards this is the next most-important policy. But, this is not as easy as we would hope. When we talk to government agencies about publishing data, their concerns revolve around privacy, security, and legal contracts with software vendors (see #1 and #2). The Oklahoma Open Records Act mandates that almost all government information be made public, but there’s a difference between a photocopy of a meeting agenda and publishing gigabytes of data about citizens in a machine-readable format.
It’s also hard to know exactly what data developers want published vs. what data is the easiest to publish. However, councilor Bynum was very interested in setting up something like civicapps.org for Tulsa as a place to decide which datasets, apps, and ideas are the best for the community.