Data Recovery Case File · Portable Drives · Case 2,100 · Milestone

An Observation That Fits Two Opposite Explanations Is a Fork, Not a Weak Observation

The two thousand one hundredth file in this archive is about a single detail that cannot settle the question it appears to answer. A drive dropped yesterday — "seemed innocuous" — now undetected by either machine, where "it looks like a small part of the connectors in the socket is missing." That one observation supports two conclusions with opposite costs, and the discipline is not to guess which, but to find the action that is safe under both.

The decode: five things this case establishes, at a milestone

The first — the detail he found is a complete explanation on its own. A socket with contacts missing cannot carry a signal. That alone accounts for the drive not being detected by anything, and it accounts for it entirely: no data path, no enumeration, on any machine. Under this reading the drive is untouched, the fault is a broken connector, and the work is a repair at the interface followed by an ordinary copy. It is the cheap branch, and it is genuinely plausible.

The second — the same detail is also evidence for the opposite conclusion. A socket does not lose contacts gently. Enough force reached that connector to break metal, and that force reached the whole assembly — the heads, the platters, the bearing. Under this reading the socket is the visible fraction of an impact that also displaced something inside, and repairing the connector would produce a drive that powers up and destroys itself. Same observation, opposite branch, opposite cost.

Why the word innocuous matters more than it looks: it is the reason the second branch stays live. The drop felt minor, and drives are damaged by drops that feel minor — the head-to-surface clearance is smaller than an airborne particle, so a fall that leaves no mark on a case can still displace what matters. His own instinct, that it seemed like nothing, is exactly the instinct that makes people power the drive to check.

The third — the test that would distinguish the branches is the one that is unsafe under one of them. Repair the socket, connect it, and you learn immediately which situation you are in. But if it is the second, you learn it by doing the damage: a displaced head does its harm in the first seconds of rotation, and the answer arrives too late to act on. This is the trap in the whole case, and it is not unusual — a great many storage questions have the property that the obvious experiment is the destructive one.

The fourth, and the resolution — look for the action that is safe on both branches, not for the branch you are on. Inspecting the internals under filtered air before any rotation costs nothing under the cheap branch and saves everything under the expensive one. It is not a compromise between the two readings; it is correct under each of them independently, which is what makes it the answer. Guessing the branch is unnecessary once an action exists that does not require knowing.

The fifth, and what this site is about. A single observation that supports two contradictory conclusions is not a deficient observation. It is a fork, correctly identified — and identifying it as a fork is a real result, because it tells you that no further looking at the outside will settle it. The failure mode is not ambiguity; it is resolving the ambiguity by assumption, picking the cheaper branch because it is cheaper, and acting as though it were established.

The principle, stated plainly for the two thousand one hundredth file: when one detail fits two explanations with different costs, do not choose between them. Find the step that is right under both, and take that one first. There is almost always such a step, and on failing storage it is almost always the one that involves not switching anything on.

What he asked that deserves answering directly: whether supplying his own media is cheaper. Yes, and it is worth doing — a suitable drive provided by the owner removes the cost of supplying one and makes no difference to the work.

On the bench

Socket damage was treated as sufficient but not exclusive before any route was chosen — missing contacts fully accounting for non-detection across both hosts, while the force required to break them also reaching the head assembly, platters and bearing, so the two explanations are indistinguishable externally. Head-to-surface clearance being smaller than an airborne particle, impacts described as minor are routinely sufficient. Inspection under the laminar flow bench preceded any rotation, being correct under either explanation.

The outcome

Socket damage treated as sufficient but not exclusive, no powering permitted pending inspection, and surfaces assessed under filtered air before any rotation. Free assessment, one fixed written figure including VAT, 50% of parts and labour upfront with the balance only on successful recovery — with owner-supplied media accepted and deducted. The decode, for the two thousand one hundredth file: your one observation explains everything and settles nothing. It is a fork — and the way through a fork is the step that is right on both branches, which here means looking before switching on.

Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.