DDOD Workflow

From Demand-Driven Open Data for HHS
Jump to: navigation, search

This page explains the details around the DDOD Workflow, along with the methods used to keep track of status and progress. The DDOD Administrator monitors progress and makes updates to inform all parties -- including User, Owner and the public -- about what's happening. Transparency and visibility are important for DDOD to be effective and follow open data best practices.


States are assigned to GitHub issues. They're intended to be non-subjective, unambiguous way to indicate who and what the use case is currently waiting on. Use cases can move back and forth between states as the use case progresses.

DDOD Github issue labels.png

  • Admin documenting
  • Admin processing data
  • Admin researching

  • Waiting on admin
  • Waiting on owner
  • Waiting on user

  • Closed — Not possible
Examples of reasons for use cases to be closed in the "not possible" state:
  • Legal reasons
  • Data owner won’t commit or release data
  • Use case out of scope of DDOD
  • Closed — Moved
Items in this use case have been moved elsewhere, merged into existing use cases, or split into multiple new use cases
  • Closed — Documented
Use cases closed in the "documented" state have had the relevant data identified, clarified, or documented
  • Closed — Workaround
Use cases closed in the "workaround" state have a short term workaround that has been:
  • Identified
  • Documented
  • Closed — Solution
Use cases closed in the "solution" state have a long term solution. Long term solutions must themselves go through multiple states which include:
  • Identification of the solution
  • A commitment from the data owner for implementation of the solution
  • Actual implementation of the solution


Milestones are sequential, progressive stage gates that a use case achieves. They remain the same regardless of the type of solution or reason for marking the use case completed. Since a use case can have multiple solutions (as evidenced by the Value Delivered category), assigning a definitive milestone is sometimes subjective.

  • 0: Request by User
  • 1-A: Clarify by Admin
  • 1-U: Clarify by User
  • 2-A: Engage by Admin
  • 2-O: Engage by Owner
  • 3-O: Discuss by Owner
  • 3-U: Discuss by User
  • 4-O: Implement by Owner
  • 4-U: Implement by User
  • 5-O: Document by Owner
  • 5-U: Document by User
  • 5-A: Document by Admin

Value Delivered

The "Value Delivered" is a category assigned to a use case that describes the type of artifacts and solutions that have been delivered to benefit the user community. A use case typically has multiple value categories associated with it, since it benefits users in different ways. Even the documentation of the use case itself has benefit, as it describes a specific situation for which requested datasets can be used.

  1. Catalog
    • Identify how datasets used for specific situations
    • Describe challenges and deficiencies in current data
  2. Document
    • Document workarounds and processing steps
    • Expose previously unindexed datasets
    • Add / improve metadata
  3. Transform
    • Scrape manual sites and sources
    • Transform PDFs into machine-readable
  4. Improve
    • Add needed fields
    • Improve data quality
    • Add / improve API
  5. Release
    • Release new datasets
    • Deploy new APIs
  6. Aggregate
    • Observe trends in use cases
    • Launch wide-reaching initiatives