The reassuring thing about a backup is that you can stop thinking about it. The dangerous thing about a backup is exactly the same.
There is a second drive. A spare connection. A little green tick. Somewhere, somebody has anticipated the disaster, and the rest of us can get on with forgetting our passwords.
Then the ordinary world reaches into the digital one and reminds everyone that a system diagram is not a guarantee.
On September 21, communications problems at an FAA facility in Philadelphia disrupted air travel across the northeastern United States. According to the Associated Press’s September 22 report, a primary communications line failed and a backup fibre-optic line had been severed during construction. Operations resumed, but disruption continued into Tuesday.
Those are reported circumstances, not a complete technical investigation. They do not establish that both connections shared the same route, or that one accident caused both failures. The distinction matters: a useful lesson should not require inventing a more satisfying disaster.
What the episode does puncture is the comforting idea that having a backup settles the question. It only starts the next one: what happens when you actually need it?
Two copies, one problem
Air-traffic communications and a family photo collection are obviously not equivalent. Nobody should design an airport around advice intended for an external hard drive. But the word “backup” can hide an assumption in both settings: that the alternative will be available when the original is not.
Consider a hypothetical home office. There are two copies of an important folder, both on drives sitting beside the computer. That arrangement might be useful when one drive fails. It is less impressive if the entire bag containing them disappears. The number of copies has not changed. The kind of failure has.
Or imagine a small business with a beautifully documented recovery procedure, stored only inside the account it needs that procedure to recover. This is not really a second way in. It is a spare key left on the kitchen counter.
The point is not that every arrangement is secretly useless. It is that protection is specific. A backup can be good at one job and helpless at another, which makes the unqualified phrase “everything is backed up” less informative than it sounds.
Even a company selling cloud backups makes that distinction. Akamai’s product guidance says its service complements a broader backup strategy when paired with an additional off-site copy. It also recommends testing restores and documenting the recovery steps. Its own description separates having recovery points from having a complete recovery plan.
That is a refreshingly unmagical way to talk about the cloud. The reassuring icon is the beginning of the conversation, not the end.
For a household, the useful question is not “Have I purchased enough technology?” It is “Which loss am I trying to survive?” A broken device, an unavailable account and a damaged home are different problems. Spending more without identifying the problem can produce an expensive collection of things that are all unavailable on the same afternoon.
There is a commercial lesson here, too. Services are easy to compare by storage capacity and monthly price. Recovery is harder to put on a shopping card. How long will it take? What must still be working? Can a person understand the instructions under pressure? Those are not glamorous features, but they are the ones that decide whether the promised reassurance becomes useful.
The boring part deserves a demonstration
The response to this week’s outage already includes promises of better infrastructure. Reuters reported on September 22 that Transportation Secretary Sean Duffy described a replacement air-traffic system with multiple telecommunications routes into FAA facilities, intended to eliminate single or dual points of failure.
That is an official ambition, not proof that a future system cannot suffer disruption. The valuable follow-up is how those promises will be tested, and what happens when the tests reveal something inconvenient.
There is a fair counterargument to all this backup scepticism: independence costs money, and complexity creates work of its own. A household does not need to operate like a national infrastructure provider. A small business cannot maintain an unlimited collection of duplicate systems. The sensible target is proportionate protection, not the fantasy of never being inconvenienced again.
Nor should an interruption automatically be treated as evidence that people should have kept operating. When essential information becomes unavailable, stopping can be the responsible response. Reliability and a willingness to continue at any cost are not the same virtue.
The consumer version of this debate can be much less dramatic. Ask for a demonstration. Restore a non-essential test file without overwriting the original. Check whether the recovery instructions are accessible separately from the thing being recovered. Find out what the service does not protect against before that exclusion becomes the most important sentence in the contract.
These are modest questions. That is part of their appeal. They replace the vague anxiety that everything might collapse with a smaller, answerable uncertainty: does this particular escape route work?
There is also something quietly healthy about making the physical world visible again. Digital convenience encourages us to think in accounts, icons and subscriptions. Maintenance asks us to think about places, people, cables and time. Neither view is complete without the other.
The best backup is not necessarily the fanciest second thing you own. It is the alternative whose limitations you understand, whose recovery you have tried, and which remains useful when the first thing lets you down.
A second boat is reassuring. Knowing it is not caught behind the same obstacle is better.
