Create a concept note for a Omang --> OpenCR Batch Pipeline

[pmanko]

We will use the Omang system to populate the Client Registry with patient identities using a batch process that would be periodically run against the Omang DB.

The Omang system will be accessed using an SQL interface, so we will likely have access to a view that has a row for each person, along with identifier and demographic information.

We need to decide:

• how we will define a possible subset of persons that we want to load into OpenCR (initial load)
• how we will filter subsequent updates of OpenCR to only include newly created or updated patients (subsequent loads)
• where we will stage the data for loading into OpenCR
• where we will host OpenCR, since real patient data cannot be stored on our demo AWS instance
• how we will test this pipeline before going live - perhaps a duplicate Omang schema with some fake data?

We should draft a possible approach as a concept note for refinement and discussion.

@tkgakge @miltonjohane Let me know if I missed anything, and feel free to update this issue as you see fit.