Data Dictionary: Identify objects to be added to the data dictionary and the format of the dictionary content

Capturing some of the notes from here. This issue is a discussion point for what we want for our data dictionary and to define the content of the dictionary. This is not an implementation issue.

Outcomes

• Identify which objects will be included in Data Dictionary MVP
  • Tables
  • Columns
• Format of DD MVP

Objects

Tables and columns can have comments, which we are currently targeting as the location for the data dictionary definitions.

Format

JSON key	Level	Description
owner	table and column	For tables where there is no clear owner either use an array if there is a clear subset of teams that own the table, or simply indicate "all" for tables that are commonly used across all teams/applications
description	table and column	A clear description of the intent of the object (e.g. column, table)
data_classification	column	How the content of the column should be classified according to our Data Classification Standard. Appropriate values are red | orange | yellow | green

Suggested format

issues - comment: {"owner" => "group::project management", "description" => "All your issues belong to us", "sensitive"=>"false"}
issues.id - comment: {"owner" => "group::memory", "description" => "Primary key sequence used as database-internal reference only", "data_classification" => "green"}
users.email - comment: {"owner" => "group::distribution", "description" => "User Email address", "data_classification" => "orange"}

Original notes &1 (comment 398083609)

Overview of PostgreSQL comments https://www.postgresql.org/docs/9.6/sql-comment.html