BIP85: replace Base64 by Base85 in PWD BASE85 section

[ethicnology]

Replace Base64 by Base85 encoder in the Pwd Base85 section, and use consistent capitalization.

[jonatack]

cc @akarve @scgbckbone @nvk @ethankosakovsky for feedback, this BIP is deployed by a few projects and is in final status (#1676), so no potentially breaking changes may be introduced.

[akarve]

@ethicnology in this spec the derived entropy, e.g. on both of the applications you commented on, is of fixed length. in what context would this be variable? there is trimming for the password length but that is explicitly post entropy. if you want truly variable password length and dictionaries (allows length of 1) maybe the new dice app is what you want?

[akarve]

@jonatack this particular change i'm asking for clarification on; so far i don't see the need. as for backward compatibility we now have a semver log. is it your opinion that even if the community adopts a major semver change that it needs to backward compatible or that said major semver change belongs in a clean BIP?

[jonatack]

as for backward compatibility we now have a semver log. is it your opinion that even if the community adopts a major semver change that it needs to backward compatible or that said major semver change belongs in a clean BIP?

Hi @akarve, yes, see #1600 (comment). I agree that breaking changes (if any) should first be discussed on the mailing list and with the projects that have already implemented the BIP, particularly for a BIP with final status, then probably go into a new BIP that replaces the previous one. Semver would be for non-breaking changes.

[akarve]

Hi @akarve, yes, see #1600 (comment).

ok this is a topic for another day. i’ll just point out that the situation leading to the above reasoning has since been addressed:

Furthermore, the BIPs process currently has no way to distinguish BIP versions and the way that has been implemented here is confusing.

[jonatack]

Furthermore, the BIPs process currently has no way to distinguish BIP versions and the way that has been implemented here is confusing.

Could take a look at BIP3 and propose any improvements there you have in mind.

[scgbckbone]

NACK

it's always 64 bytes as it uses whole hmac-sha512 digest

[ethicnology]

it's always 64 bytes as it uses whole hmac-sha512 digest

I understand, I was confused with the password length.

I've amended my commit to replace only the Base64 by Base85 in the PWD BASE85 section

[scgbckbone]

@ethicnology could you also unify capitalization on: base64 -> Base64 & base85 -> Base85

[scgbckbone]

ACK 3e9befd

[jonatack]

ACK 3e9befd

[jonatack]

cc @akarve for feedback or sign-off

[akarve]

Approved