Thanks again to those who joined. I went through the notes and tried to pull out a set of takeaways, itemized at the top of the notes and pasted below.
Next week many of us will be at the Inform USA conference, in which these converastions can continue informally.
Is there interest in re-starting the conversation in June? Say June 3rd or 10th at 11aEastern? If so, what should the agenda be?
see below and chime in – thanks!
Takeaways:
Roles:
- Utility: responsible for aggregate, publish, and quality control. May have specific front-line stewardship responsibilities in addition to bottom-line / umbrella responsibilities; may also play role of curator and auditor for other stewards.
- Steward: responsible for a specified level of data management of a given set of resources.
- Auditor: responsible for assessing the quality of data managed by a given set of stewards, to ensure compliance with standards. (performed by Utility
- Curator: responsible for ensuring consistency in subjective elements of resource records, especially category/taxonomy
There should be designated stewardship responsibilities for the ‘core’ part of the record. That said, taxonomies might benefit from being managed (curated) centrally.
For aggregated vs unbundled service information – i.e. Programs that might involve multiple subsidiary services – Perhaps bundled program record is shared broadly, and discrete service records are kept locally and/or provided for a fee.
^ Question: Can modern transformer tools help accommodate a both/and balance between loose and strict? So that systems can have it both ways with tooling to automate bundling/unbundling.
Style guide : data standard :: style template : exchange profile
in order to get alignment on complex orgs, develop a ‘template’ for certain kinds of organizations (like gov agencies) to specify how information about them should be structure.
In order to get alignment on specific data point to share for specific purposes/users, develop an exchange profile – “Export profile” or “import profile” – that specifies particular fields for particular purposes, and sets up the crosswalking to be automatable.
Who should get paid for what?
- Locals / domain experts / stewards should get paid for managing records –
- on a per-record basis?
- Regional/state-level utility should get paid for aggregating comprehensive data, quality assurance, etc.
- Producing “opinionated” data sets for specific consumers
- Participation in cooperative processes i.e. governance activities
- Some local work (non-standard contributions, unstructured input) can be contributed by community partners who might not get paid but can benefit in nonmonetary ways (training, support)
- CBOs should get paid for providing services! This is important overall; but it may or may not be relevant to data management strategy.