Contribution criteria: Design goals, inclusion requirements, and paths to core.

[justlevine]

What problem does this address?

As we get closer to an initial working release, it is important we work to ensure we are following the official Criteria for Creating or Migrating Repositories under the WordPress GitHub Organization.

Of particular consideration is that this isn't just a canonical plugin, but a collection of Experiments that - in theory at least - can one day graduate to core. So while contribution guidelines are always important, it also influences our architectural decisions.

What is your proposed solution?

As I see it, we need to decide on and then document / update the existing design goals to broadly cover:

• When we're deviating from WPCS and WordPress core patterns, and why.
• How an Experiment gets approved for inclusion in the plugin
• What roughly does the graduation from Experiment to core look like

[jeffpaul]

What roughly does the graduation from Experiment to core look like

IMO "we", the WP AI team, should NOT be deciding that. It should follow, more or less, the feature merge criteria already defined within the Core handbook. If "we", again the WP AI team, feel anything there needs to be updated, then that should be a conversation we have with the broader core team.

[justlevine]

What roughly does the graduation from Experiment to core look like

IMO "we", the WP AI team, should NOT be deciding that. It should follow, more or less, the more or less, the feature merge criteria already defined within the Core handbook.

I'm talking about the processes from our side to say "yeah lets cut a PR and get the code ready for core". In practical terms, it also plots the deviation between us and WPCE/Core patterns e.g. the amount of acceptable/expected work to decouple (or introduce) our abstractions, migrate from DI or OOP, etc.

[jeffpaul]

When we're deviating from WPCS and WordPress core patterns, and why.

I'd like to minimize this to zero if possible, at least over the next ~2 months.

Everything else in this issue is worth discussing, but I think is more a problem for future-us and something I'd like to not worry about until January while we focus more on getting as much polished and into the plugin by DEC 2nd. Happy to have folks continue the discussion here, but I'll likely need to focus elsewhere in the short-term.