BIP-XYZ: Standardized emoji based wordlist for future apps

[ovrdos]

Emoji Seed Word List Pull Request

Purpose

The purpose of this PR is to define and discuss the proper standardized definitions an official emoji seedlist.

Research

All research and discussions on the finalized standard are included in the README on https://emojiseed.com.

Community

The EmojiSeed community is just starting if you would like to join as a contributor or a watcher feel free to speak out.

Learn more about the specification and community at:

• https://emojiseed.com/
• https://emojiseed.com/discussions

[Adriano1324]

In this review, since there are no code changes, I will refer to the code in the following repository:
https://github.com/emojiseed/bip-emojiseed/tree/testing

If a simple list of word-to-emote-pair correlations is considered sufficient, then there isn’t much to do. However, for easier usage and better maintainability, it would be beneficial to implement this as a library that provides:

An API to retrieve pairs of emotes

Support for importing translations, custom sets, and themes

Functionality for searching through the data

Regarding the website used to view emojis:

after uploading csv it only displays question marks instead of emojis

It lacks filtering options — ideally, it should allow filtering by meaning or selecting specific emotes for which we want to view the corresponding pairs

text is too small

Inside of tests is node_modules directory which should be added to .gitignore and removed

[murchandamus]

This strikes me as a bad idea. Skimming the provided emoji list, there are a number of symbols that look quite similar, and especially since emojis are rendered differently on various platforms, this approach may be a lot less reliable than necessary.

I could imagine that a drastically smaller set of distinct emojis, in combination with a higher count of symbols per “word” could be a working solution, but I’m not sure whether this would actually be serviceable or a net-improvement.

Alternatively, instead of replicating the word list approach, I would imagine that a sequence of a dozen or a score emojis picked from a smaller distinct set could be sufficient to directly encode a master secret.

The document would also need to be extended to meet editorial standards.