Mention what type of data goes into the secret fields

[adamryczkowski]

After 15 minutes of reading the documentation, I am 50% sure that each blank for the secret should contain a single letter out of the alphabet of 25 distinct tokens (with X and J merged together), although the gap is large enough for a short word.

I understand that for the author this is super-obvious, but me, as a causal visitor who is not familiar with Shamir's secret sharing (and uses this work to try to familiarize oneself with the concept), this piece of information is missing with no place even to deduce it.

If the ambiguity is there on purpose because e.g. there are several operating modes the psst can operate - e.g. a secret comprised from 25-level tokens, or just digits, or maybe some other type of data - please provide an example. That can be in a form of a photo of the filled in form. Even better yet would be an accompanying video of the process of filling in the form and then recovering the secret.

And thank you for sharing this! :-)

[Sjlver]

Thanks for the feedback. You're absolutely right, this is a place where the documentation should be improved. An video tutorial would be great too, but time-consuming to make.

The short answer is: You can use the space in any way you like. The numbers 1-24 are there to add a bit of structure that might be helpful, but feel free to ignore. A few suggestions:

• If you use psst to backup a password with 24 characters or less, you can put 1 character (= 3 0-4 digits) into each field.
• Some people might use psst for passphrases, which often have 12 or 24 words. In this case you can use one or two fields per word. They would be quite full, since a five-letter word needs 10 digits.

We recommend that you document the chosen approach in the description of the share, so that there is no ambiguity during the reconstruction of the secret.

If you end up using psst, feel free to send me some photos! Just take care to not accidentally include secret data in the photo.