8. [Repositories] Ontology Data Pipelines28. Exercise Summary

28 - Exercise Summary

This content is also available at learn.palantir.com ↗ and is presented here for accessibility purposes.

The image above depicts the part of the writeback infrastructure you’ve configured so far. You’ve synchronized backing datasets into the object storage service, built Ontology-aware applications, configured writeback datasets for your object types, and scheduled builds for those writeback datasets. What you’ve not done yet is create Actions that write edits back to the storage service. This is the subject of multiple tutorials in the APPDEV training track.

✅ What you built

  • Writeback datasets for your flight alert and passenger object types.
  • A writeback dataset for the join table backing your passenger ⟷ flight alert link type.
  • A schedule for your writeback datasets.

✅ What you learned

  • Once writeback is enabled for an object type, the corresponding writeback dataset should be considered the canonical data source for further Foundry work rather than the original backing dataset.
  • If you want to enable link modification actions on many-to-many link types, you’ll need to add a writeback dataset to the backing join table.
  • It is a best practice to only allow edits via Foundry Actions, though other configurations are possible (see documentation for more details).