When designing SAP on AWS, the book (quoting Thomas Sowell, "there are only trade-offs") says architectural trade-offs collapse into a small fixed set of categories. Which categories, and what does "trade-off" mean here?

An SAP architect secures the network perimeter heavily but leaves the EC2 OS, the DB, and IAM relatively open, reasoning the firewall is enough. Which Security-pillar principle is being violated, and why does it matter?

A reporting SAP system that was never reviewed under the SAP Lens has just been redesignated business-critical and now needs HA — but it's already in production. Is it too late to run an SAP Lens review, and where do these reviews pay off most?

Why does moving SAP transports in small weekly batches beat a single large monthly transport window, even though weekly means more change events?

How is a Sustainability-pillar goal like "less compute/storage per SAP transaction" different from a Cost Optimization goal, given both push you to use fewer resources?

You can deliver the same total SAPS with either three large SAP application servers or nine smaller ones. Why is the larger fleet of smaller servers the more resilient choice?

You can technically mount Amazon S3 to an EC2 instance. Why does "mechanical sympathy" tell you NOT to put database files on it?