TokenScript repo

When will it be used

A user obtains a TokenScript in two ways.

1. If the user purchases or obtained a token through a website, the TokenScript for that new token is shipped from that website[^1.

2. If the user needs a TokenScriptn without going through a dapp website, (e.g. he receives a token from a friend), the user-agent (wallet) tries to find the TokenScript for that token by accessing a repo server, sourcing data from this repo.

This repo can be decentralised with the rise of reliable decentralised storage.

Submit a TokenScript to this repo

When you deploy your Tokenscript, please also send a copy to this repo in the form of a new Pull Request. By doing so:

1. If a user gets a token without going through any of the dapp websites with your TokenScript deployed, they can still get a copy from here.

2. When we upgrade the schema, we know who to talk to for updating existing TokenScript.

When you submit the TokenScript, create a new directory following the convention as specified.

When a user-agent (client) obtains a TokenScript from this repo

• A user receives a token directly.
• A user scans a QR code or opens a link that represents a token-holding smart-contract.
• A user gets a "MagicLink" which allows him to redeem a token.
• A user recovers a key backup, and the user-agent discovers that he/she owns a few Token.

Check user-agent behaviour for how a user-agent is supposed to communicate with a repo server.

Will namespace and schema always be updated at the same time?

Likely so in 2019 and perhaps 2020 as well. In the first two years, TokenScript schema and namespace are versioned in the same way. A user-agent only works with TokenScripts whose schema and namespace are both supported by the user-agent.

Beyond 2020, a user-agent would work with any TokenScript whose namespace it supports, even if its schema is not supported. By such an arrangement, TokenScript schema can evolve at a faster pace than its namespace, by adding new elements and not removing old elements; but it leaves room for mistakes, resulting existing Tokenscript becomes invalidated. We think it's a fair trade-off.