Prioritizing Open Data Sets
Download as PPTX, PDF0 likes306 views
The document provides 3 rules for prioritizing open data sets: 1) Look at past public requests and what other governments publish to identify in-demand data; 2) Target policy objectives like health, accountability, and transportation; and 3) Partner with others when possible and follow common data structures to maximize usefulness and access, but avoid exclusivity and move quickly to establish standards.
Prioritizing Open Data Sets
- 1. Prioritizing Open Data Sets3 simple rules
- 2. Rule 1: Look Backward, Look SidewaysLook Backwards:Most governments know what data and information has been heavily requested by the public in the pastPast requests reveal both latent demand and potential communityThis is a good place to startLook Sideways:Look at requests other governments make public, there are trendsExamples of data sets that regularly top the list: procurement, budget, crime, transit and transportation data
- 3. Rule 2: Target Policy ObjectivesRelease data that will advance specific policy goalsHealthAccountability/anti-corruptionEducationTransportEnvironment
- 4. No one visits your websiteThis is Vancouver Coastal Health’s Restaurant Inspection site.No one visits it.
- 5. Repeat, no one visits your websiteThe only time anyone searches for restaurant inspection data.....is after they’ve been food poisoned
- 6. e.g. Health ObjectivesWhen citizens see restaurant inspection dataConsumer behavior drives compliance:Well rated restaurants experience more businessBadly rated restaurants experience less businessConsumer behavior and compliance create savings:In Los Angeles County, Emergency Room Visits due to food born related illnesses declined by 16% in year one, 6% in year two, 3% in year threeFocused transparency can improve health outcomes and lower costs
- 7. Put the data where the citizen is
- 8. Rule 2: This is your goal
- 9. Rule 3: Partner when you must, Follow when you canOpen data in structures that are common across multiple jurisdictions is much, much more powerfulThe power of a common structure: the General Transit Feed Specification allows access to transit schedules in 160+ cities in dozens of languages in Google MapsCopy others structure whenever you can
- 10. Rule 3: Partner when you must, Follow when you canRule 3a:Help foster cross jurisdictional standards by copying other people’s data structuresRule 3b:When there is no standard, find private sector or community actors who want to consume your data, and get them to help you create the data structureTell the world about your structure!
- 11. PartneringCreating a new standard requires scale, access to a large user base, and an ability to compromise. E.g. For Transit, Portland engaged Google4 Do’s and Don’tsDON’T start by forming a standards committee, we don’t have 10 years to spend coming to agreementDO find a partner with scale and market penetrationDON’T promise exclusivity to that partnerDO create a governance model after the standard succeeds