“TODO comments don’t need to get done in order to be valuable.”
Alpert makes the case that TODO comments are not task lists but preserved context from the original author’s brain. A good TODO explains the edge case the author knew about but didn’t handle, and that’s more valuable than actually fixing it in most cases. The distinction between a TODO that says “write the rest of this file” and one that documents why the code works this way despite looking wrong is the difference between laziness and institutional knowledge.