SURVIVAL APPSWaste and redundancy: what to throw out and what to keep in duplicate
Anyone who lives off salvage learns an uncomfortable rule early: the same thing can be a lifeline or dead weight, and sometimes the only difference is where you keep it and why. I spent one afternoon emptying a drawer of cables that no longer power anything, and that evening buying a second disk identical to one I already had. It's not a contradiction. It's the trade.
Waste and redundancy are not the same word
People mix them up constantly, because from the outside they look alike: in both cases you have two of something, or you're holding on to what you could throw away. But they're opposites. Waste is a resource that goes out and never comes back: energy burned by a device left on for nothing, storage taken up by stuff you'll never look at again, money leaving every month for a service you don't use. Redundancy is a resource spent on purpose for the day the first copy betrays you: the second disk, the spare battery, the procedure written on paper for when the screen is dead.
Survival culture has a motto for this: one is none, two is one[1]. A single thing, when it fails, leaves you at zero. The same mindset, applied badly to computing, produces monsters: ten apps that do the same thing, three backups that are really the same folder copied by hand, a digital closet nobody will ever reopen. The right question isn't "do I have two?" but "does the second copy save me, or does it just weigh me down?".
Computing waste you pay for without seeing it
The nasty thing about digital waste is that it's invisible: it doesn't clutter the drawer, it doesn't make noise. That's why it piles up. Here's what I hunted down by looking at my own machines:
- IDLE ENERGY — always-on devices that work 0% of the time: a mini-server, an old network printer, a set-top box in standby. Constant watts multiplied by 8,760 hours a year: that's where the electricity bill hides
- ZOMBIE STORAGE — backups that were never compressed,
Downloadfolders weighing tens of gigabytes, photos duplicated three times across three devices, Docker containers and images nobody runs anymore. They take up space and they get backed up too, multiplying the waste - GHOST SUBSCRIPTIONS — cloud paid for a terabyte when you use twenty gigs, tools signed up for "just to try" and never cancelled, two services doing the same job. Waste that costs you cash, not just resources
- DEAD CODE AND DEPENDENCIES — functions nobody calls, libraries imported and never used, feature flags switched off for a year. Not just weight: extra attack surface and time lost dealing with them at every change
- FAKE REDUNDANCY — the worst, because it fools you. Three copies on the same disk aren't three copies: they're a single failure that takes them all out together. It looks like safety, it's waste in disguise
The redundancy you should keep (and look after)
Once the waste is cut, what's left is the duplication that earns its place. Good redundancy has one defining trait: it's independent. Two copies that can fail together are worth one; two copies that fail for different reasons are really worth two. The classic backup rule spells it out in black and white.
| Principle | In the physical world | In the digital world |
|---|---|---|
| Three copies | Original + two spares | 3-2-1 rule: 3 copies of the data[2] |
| Two different media | Battery + solar panel | Local disk + NAS, never the same volume |
| One copy off-site | A stash at someone else's place | Encrypted offsite/cloud backup |
| The analog fallback | Paper map, hand-crank flashlight | Passwords on paper, printed procedure |
Redundancy also has to be tested: a backup you've never tried to restore isn't a backup, it's a hope. It goes for a file as much as for the generator you never start — the day you need it is not the moment to find out it won't turn over.
> storage: 42 GB in duplicates and dead caches
> energy: 3 always-on devices, real usage < 2%
> subscriptions: 2 overlapping services, 1 never opened
> verified independent backups: 1 out of 3
> verdict: trim the fat, lock down the real reserves
An audit you can do in one afternoon
You don't need an elaborate method, you need to actually do it. I've boiled it down to four steps, the same for the cable drawer and for the disk:
- MEASURE — what's really consuming? A plug with a wattmeter for energy,
du -sh *for storage, your bank statement for subscriptions. Without numbers you're guessing - DISTINGUISH — for each item: waste (goes out, never comes back) or redundancy (saves me)? If it's redundancy, is it independent of the first copy, or does it fall together with it?
- CUT — waste goes without regrets: switch it off, cancel it, delete it, remove the dead dependency. One month of trial before you throw it out, then it's gone
- HARDEN — real redundancy gets made independent and tested. Move the second copy to another medium, try a restore, print the fallback
Frequently asked questions
What's the difference between waste and redundancy in computing?
Waste is a resource that goes out and never comes back (energy, storage, unused subscriptions). Redundancy is a resource spent on purpose for the day the first copy betrays you — a second disk, an independent backup.
What makes redundancy "fake"?
Having several copies that can fail together, like three copies on the same disk: it looks like safety but a single failure takes them all out. Real redundancy is independent, on a different medium.
How do you audit computing waste in one afternoon?
Four steps: measure what's really consuming, tell waste from redundancy, cut the waste without regrets, and harden and test the real redundancy by moving it to a different medium.
The cable drawer is lighter, and the second disk spins in another room. I threw things out to make room for what, if the need comes, I don't want to have as a single copy. The thin thread of choice.
Notes
- One is none, two is one — a widespread maxim in survival and military circles: relying on a single example of a critical item is the same as not having it, because a failure leaves you empty-handed. Hence the habit of reasoned stockpiling (not hoarding). ↑ back to text
- 3-2-1 rule — a backup best practice: at least 3 copies of your data, on 2 different types of media, with 1 kept off-site. It's meant to ensure that no single event (failure, theft, fire, ransomware) can wipe them all out together. ↑ back to text