daranjjohnson
commented
6 months ago I find documentation is best if it can be found at the source. However, I rarely see this done in the wild. For relational database systems, if there is documentation (data dictionary), it is generally in a PDF/Word/Excel file. This is sub-optimal since it can easily get lost or forgotten.
There is a better way - for each object (table, column, view, etc.) there is a comments field that is almost never used. That comments field is the best way I have found to add and maintain documentation (data dictionary). Additionally, these fields are queryable. An official Data Dictionary can be created by creating a view with these fields and using a reporting platform (Tableau/Looker/Power BI) to create a Data Dictionary report. This would be easy to maintain as long as every time the table, column, view, etc. are created/updated the comments fields are also updated.
Crosspost: Why you’re closer to data documentation than you think | Emily Riederer
Writing is thinking; documenting is planning and executing. In this cross-post with Select Star, I write about how teams can produce high-quality and maintainble documentation by smartly structuring planning and development documentation and effeciently recycling them into long-term, user-friendly docs
https://www.emilyriederer.com/post/docs-closer-than-you-think/index.html