** I can only speak for the Trezor 5 BTC-only version. **
I really love your products and I’ve been a buyer since day one.
But in my opinion, it’s quite disappointing that after all these years on the market, this product still lacks some really great features I’ve been looking forward to for years.
If you take a look at your competitors, most of them support all of these functions, and I don’t understand why you do not.
Multisig support: It’s still not possible to register a multisig setup, with all its cosigners, on the Trezor.
Master Fingerprint: There is no option to see the fingerprint of the wallet you are on - not in the Trezor Suite App nor on the device itself. It would be nice to have at least a toggle option to turn it on or off. This helps a lot if you are using different passphrases and need fast identification.
Calculating the Seed Phrase checksum word: As we know, if you have a seed of 12 or 24 words, the last word is a checksum calculated from the 11 or 23 words before.
So, there is a limited set of words that can validate the seed. If you enter 23 words into the Trezor, it not only gives you the valid words to fill in, it lets you choose from any word, only to tell you afterward that it’s an invalid seed phrase.
Why not calculate the possible last words and only give these options?
This would be a really nice feature.
For example, if you created your seed by dice rolls or in a different way, you could use the Trezor to find a valid last word for you.
It could also prevent mistakes when entering your seed phrase during recovery.
I would really like to see these features implemented in a future version of the Trezor.
And I don’t see a good reason why they shouldn’t be.
Do you think your customers would be overwhelmed by these features? They are smarter than you think! And even if so, it’s just a feature that you can choose to use or not.
Such feature would do the exact opposite of “preventing mistakes”. It would actively hide mistakes. The user would assume that their error is in the last, checksum word – whereas in reality, the error is 11x more likely to be in one of the other words. An error-corrected seed in this manner would be the wrong one in 11 out of 12 (or 23 out of 24) cases, and nothing would tell the user this. They would just get an empty wallet.
There could be a dedicated feature for dice rolling your seed, sure. But that would need to be its own thing to not mess up the UX of the “normal” recovery.
Thank you for your response, and I would like to share my thoughts on this with you.
On this point, I do not agree with you.
Let’s assume someone wants to restore a seed with 24 words. They have all 24 words saved and now enter the first 23 words into the Trezor.
Now, for the 24th checksum word, the Trezor offers a selection of 8 words. (Because with a 24-word seed, there can only ever be 8 different checksum words.)
If none of the 8 suggested words matches the previously saved 24th word of the seed, it is obvious that one of the preceding words must be incorrect.
If this approach had such a high potential for errors, why have all other competitors implemented it as standard?
But if you feel more comfortable with it, of course, you can also implement a standalone function that allows you to calculate the last checksum word.
a more oblivious user (and trust me, most of them are more oblivious) will assume that the wallet is broken and not accepting the “correct” last word.
whereas when you allow them to enter their actual word, and then say that the seed-as-a-whole is invalid, they can re-examine whether they made a mistake anywhere in the seed
this is also much less interesting for 12-word seeds where there are 128 options for the last word.
they have? which ones specifically? if they’re doing it, I’d love to take a look at the overall UX
I agree with you that it doesn’t make sense to display all 128 possible checksum words for a 12-word seed.
However, for example, you could initially grey out all impossible letters and, with each letter entered, quickly narrow down the available options. On average, after 2-3 letters, there should only be a small number of valid words left.
Yes, most competitors have implemented this by default.
I have personally tested this with the following hardware wallets:
BitBox02, Jade Plus, Coldcard MK4, SeedSigner.
I also know that it works with the following devices:
Jade, Coldcard Q.
Have you already had the opportunity to take a look at the user interface of one or more of the other devices?
I would be interested to hear your opinion on it and what you think about the implementation of the checksum word calculation there.