Formosa aka BIP 450 has already been posted about twice on here (#1498637, #1494889), but it didn't get much attention and seeing as the name doesn't exactly explain what it does, I thought it'd be interesting to discuss.
BIP 450 proposes a new encoding layer for BIP 39 seed phrases that turns them into 4 little story chunks or "meaningful, themed sentences."
BIP-0039 maps each 11 bits of entropy to one word drawn from a single 2048-word list, Formosa maps each 33 bits of entropy to a short themed sentence built from several smaller, syntactically-typed wordlists. The sentences carry grammatical structure and semantic coherence, substantially improving memorability while retaining all cryptographic properties of the original scheme.
The proposal is fully forward- and backward-compatible with BIP-0039: BIP-0039 is itself a Formosa theme, and seed derivation re-encodes any Formosa mnemonic through the BIP-0039 English wordlist before applying PBKDF2, so existing keys and addresses are preserved.
The proposal seems like it has been worked on for quite a while, and there are mailing list discussions about it (here and here and it was recently assigned a BIP number.
Assuming that a scheme like this doesn't weaken your entropy,
Would you prefer a seed phrase that was easier to remember because it was a set of sentences?Would you prefer a seed phrase that was easier to remember because it was a set of sentences?
I'm not sure for myself. I remember that, in my innocence, I once memorized the seed phrase of one of my early wallets. Twenty-some wallets later, I couldn't tell you a single word from that original wallet.
Additionally, the developer behind Formosa (Yuri S Villas Boas) came up with a neat way of using Formosa to defeat side channel attacks:
This in turn enables the companion project Mooncake, which renders each Formosa category as an on-screen table. The words themselves stay in their alphabetical positions in the table (so the user can locate them visually); what is randomized per input session is the indexation, i.e. the labels (numbers or short codes) that the user must type to designate a given cell. The user therefore enters a sequence of session- specific indexes rather than the words themselves.
I prefer the standard random seed words and later I can build my own sentences with those words. As we did that funny experiment here on SN: #1260936
'Formosa' is a portuguese word, the best english translation here would be 'Comely' (someone who's easy on the eyes).
Not bad! The name does suit the project. I assumed it was connected to the Island of Formosa somehow.
It is, in the sense that it was also Portuguese sailors who gave Taiwan’s main island that name, presumably because the Ilha Formosa was easy on their eyes. ;)
I was just about to say that! The most common meanings are pretty, beautiful, stunning, and elegant.
Learning new things every day. I think my Big Trouble in Little China childhood mind somehow assumed Formosa was a Taiwanese term corrupted by English maps.
To answer your question: absolutely. Standard BIP-39 words are great for machines, but terrible for human brains over long periods. A structured narrative pattern means you can reliably store a backup in your mind without it decaying after twenty wallets.