Exercise 3: Why a "Last Reviewed" Date Is Honest but Not Sufficient — Possible Solution ==================================================================== WHY IT'S GENUINELY HONEST ------------------------------ A "last reviewed" date, when accurate, tells the reader something real and useful: whether the document has been looked at recently at all. Per this chapter, it's "an honest, immediate signal letting a reader judge how stale a document might be, even before discovering whether it's actually wrong." A document last reviewed three years ago is legitimately more suspect than one reviewed last month - the date itself isn't misleading. WHY THE DATE ALONE DOESN'T GUARANTEE ACCURACY ------------------------------ Per this chapter's own warn-box, "a document reviewed recently could still have been reviewed carelessly - a rubber-stamped date bump with no genuine scrutiny behind it." The date only records that a review event occurred, not what quality of review actually happened during it. Someone could open a document, confirm it still exists, and update the timestamp without ever actually checking whether the hostnames, procedures, or values it references still match current reality. WHAT COULD MAKE A RECENTLY-REVIEWED DOCUMENT STILL WRONG ------------------------------ A reviewer under time pressure, unfamiliar with the specific system the document covers, or simply treating the review as a compliance checkbox rather than genuine verification, could update the date without meaningfully checking anything - the same failure mode as Chapter 4's own warning that a runbook missing rollback implicitly assumes success, except here the review process itself becomes the thing that quietly assumes accuracy rather than verifying it. WHY THIS ISN'T A CONTRADICTION OF THE DATE'S OWN VALUE ------------------------------ The date isn't dishonest or useless - it genuinely does distinguish "reviewed recently" from "never reviewed at all," which is real, useful information. It's specifically insufficient as the *only* safeguard, because it measures whether a review event happened, not whether that event actually caught anything. WHY THIS MATTERS FOR HOW THE DATE SHOULD ACTUALLY BE USED ------------------------------ The date works best as one honest signal among the chapter's own several countermeasures (ownership, a defined cadence, updates tied to change processes) rather than as a stand-alone guarantee - a recent date should raise confidence, not eliminate the need for the other safeguards to also be genuinely working. WHY THIS WORKS AS AN ANSWER ------------------------------ It explains what real, honest information the date does provide, explains specifically what it fails to guarantee (genuine scrutiny during the review, not just the review's occurrence), and explains why this makes it one honest signal among several rather than a sufficient safeguard on its own.