1.1.1. From Click-Ops to Code: The Core Shift
💡 First Principle: Every manual action in a cloud console is a decision that lives only in someone's head and the provider's database — never in a place your team can see, review, or replay. Moving that decision into code is what makes infrastructure knowable.
"Click-ops" — provisioning resources by clicking through a web console — works fine for one person experimenting. It collapses at team scale because there is no record of why a setting was chosen, no way to review a change before it happens, and no guarantee that the staging environment matches production. When the person who built it leaves, the knowledge leaves with them.
Infrastructure as Code replaces those clicks with declarative files committed to version control. Now a change to a firewall rule is a pull request: visible, commented on, approved, and traceable. If production breaks, you can git blame the exact change. If you need a second identical environment, you copy the code. The shift is not primarily about speed — it is about making infrastructure behave like software, with all the discipline that brings.
⚠️ Exam Trap: The exam may describe Terraform as "configuration management." Terraform is a provisioning tool — it creates, updates, and destroys infrastructure resources. Tools like Ansible, Chef, and Puppet specialize in configuring software inside already-existing machines. They overlap, but the exam expects you to know Terraform's primary job is provisioning.
Reflection Question: A colleague says "we already automate everything with shell scripts, so we already do IaC." What capability are they most likely missing that a tool like Terraform provides?
