Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Participants

  • John Raub

...

  • Erik J, Ed

...

  • Analytics
  • Andrew Rice
  • PatrickYoho
  • Shanon Kerlick
  • Jean Francois G
  • Jim McKay
  • Eric Jansson

Notes

The discussion was mostly around option set mapping, and less around identity mapping.

In the discussion, there was a central philosophical question: can you unify governance contexts, which is the central driver of data changing as it moves? There was no clear answer, but most agreed that this was possible in narrow contexts such as "rostering" but impractical for school district and SEA operational and analytical needs. 

Given  that, there will always be a need to translate data between contexts.

The complexity of this translation was discussed: mappings from more specific descriptors to generalized ones are generally simple (Grades 1, 2 and 3 => Elementary), but the opposite case is difficult (Elementary => Grade 1 doesn't work).

We discussed if the inverse case is important (general to specific) in the Ed-Fi ecosystem, and there were no compelling reasons seen to support Accordingly, the hope  was  come up with technology designs for API and other platform support for this that behaves 'logically" but does not need to fully engage the complexity, as the use cases for general => specific have not materialized.

Education Analytics also noted that they are working on a grant-funded project too produce a descriptor mapping system that uses the current operational context tables in the ODS. In disc the UI for this was seen  as aa likely determinant of success.

Background Materials

See TAG Meeting 2017-08-09

See TAG Meeting 2017-11-01