What are paper wallets?

A “paper wallet” is simply key material written down or printed out. Years ago that often meant a single raw private key. These days, people usually mean a seed phrase (often BIP39). It’s old-school, fully offline, and – crucially – deceptively easy to mess up. Done well, it can be a solid backup. Done badly, it’s the sort of setup where one small slip turns into a total loss.

The appeal is obvious: it’s cheap, there are no electronics to fail, and nothing is connected to the internet. The downside is everything around it. Printers can retain jobs, phones and cameras exist, malware exists, and “offline” sometimes means “offline-ish”. Spending from a paper setup safely can also be awkward, because at some point you have to bring that key material back into a device that can sign – which is exactly where many people leak or expose it. In practice, paper is best treated as backup material, not a daily signing workflow.

If you’re determined to do “paper” as safely as possible, you’ll usually see one of these approaches.

One approach is using an offline BIP39 tool – Ian Coleman’s Mnemonic Code Converter is the classic example – properly. The advantage is that it lets you generate and verify seed phrases offline in a repeatable way, and you can sanity-check what’s happening. The risk isn’t the maths; it’s the ops: download the wrong thing, don’t verify what you downloaded, use a machine that isn’t genuinely clean/offline, or leak via storage/printer habits, and you can defeat the whole point without realising.

Another approach is dice-based BIP39 – for example bip39-dice-style workflows. The big selling point is human-verifiable entropy: you create randomness with dice instead of “trusting the computer”. The trade-off is human error: mis-rolling, mis-recording, or messing up the conversion step. The nightmare scenario isn’t “someone steals it”; it’s “you can’t reliably reproduce it later”, which is exactly what you don’t want from a backup.

The third is something like the Glacier Protocol: a structured, paranoid-grade playbook for generating paper keys and (often) multisig, aimed at reducing single points of failure for serious amounts. The upside is the discipline – it’s a checklist-driven process with fewer “winging it” moments. The downside is complexity. It’s time-consuming, easy to botch if you rush, and it only works if you follow the process properly rather than doing a “lite” version and hoping for the best.

A good mental model is: paper can be a decent disaster-recovery backup, but it’s fragile – physically and operationally. If you do use it, slow down, double-check every word/order, avoid printers unless you truly know what you’re doing, and consider graduating to a more durable backup medium (metal) once you’ve proved you can recover successfully.