License property improvements

[merkys]

This PR addresses issues raised in #494 by:

• defining what an unknown value for available_licenses means;
• including the structure of the database among the things that are covered by available_licenses;
• introducing available_licenses_for_entries to specify licenses for database content and metadata only (without its structure).

Fixes #494.

[rartino]

I have a few proposals that I think helps improve the clarity and stay clear of things possible to be interpreted in unintentional ways.

[rartino]

I guess I was holding off for other conversions to settle; but I think the current version is good enough to merge. Thanks for the work on this!

[merkys]

Thanks @rartino. It would be nice to get approvals from more OPTIMADE adopters as this PR changes the interpretation of license properties (new in this release, though).

[rartino]

@ml-evs In the spirit of cutting down on our PR log - since you've commented on this previously - do you still think there are remaining issues, or could you consider approving it?

@merkys

It would be nice to get approvals from more OPTIMADE adopters as this PR changes the interpretation of license properties (new in this release, though).

This PR clarifies a few situations that were previously unclear, and extends what can be expressed. But as far as I see, no behavior that was well-defined with the previous PR changes by this PR. (Feel free to enlighten me if I'm missing something.)

[ml-evs]

I think the current wording is clear to me (and circumvents my need for examples). Thanks all!

[rartino]

@merkys Are you comfortable 'resolving' the two remaining conversations and merging this? (Apparently leaving conversations unresolved is now a blocker that cannot be overridden - I'm not sure how I feel about that.)

[merkys]

@rartino

@merkys

It would be nice to get approvals from more OPTIMADE adopters as this PR changes the interpretation of license properties (new in this release, though).

This PR clarifies a few situations that were previously unclear, and extends what can be expressed. But as far as I see, no behavior that was well-defined with the previous PR changes by this PR. (Feel free to enlighten me if I'm missing something.)

This is correct. I guess I tried to be extra-safe here with talking about the changes to the interpretation.

I have just marked all the remaining conversations about the text as resolved.