An ode to well-written READMEs. Detailed, but not overly so, and scannable with examples.

Life without a REPL, and how to still be able to manipulate production which even has quite a few benefits over more one-off REPL-driven operations..

On how software that survives first encounter with production tends to live longer than we think it should.

Tweaking error handling in our idempotency layer with a new transience flag.

In which we address the perennial question of when it’s appropriate to push domain logic into database stored procedures and triggers versus keeping it at the application layer.