Fix language suggesting that optional parameters "function the same way for all DIDs"

[jandrieu]

The text currently says

		Some DID parameters are completely independent of any specific <a>DID
		method</a> and function the same way for all <a>DIDs</a>.

This isn't quite correct. Some DID parameters are designed to be independent of any specific method, but in practice are not.

The same error persists in the latter sentence:

The following table provides common DID parameters that
		function the same way across all <a>DID methods</a>. Support for all
		<a href="#did-parameters">DID Parameters</a> is OPTIONAL.

These parameters are designed to function the same way across all DID methods, but since support is optional, they may not.

This was fixed in PR#311 and #315, but those are being deconstructed.

[peacekeeper]

I don't see any error here. E.g. the DID parameter "service" is independent of any specific DID method. It functions the same way across all DID methods. The fact that support is optional doesn't change anything about that, it just means that some resolvers may not implement this parameter.

[wip-abramson]

I agree with @jandrieu, this language could be clearer.

The language in #315 is good, apart from this:

The following table provides common DID parameters that SHOULD function the same way across all DID methods

Which I think is what @peacekeeper is getting at too.

These DID parameters, while OPTIONAL for a DID Resolver / DID Method to support, if supported the MUST function the same way.

A DID Method cannot, by my reading, overwrite how the service parameter works.

...

That was my understanding, but rereading the DID Parameter's section it doesn't actually say that anywhere. So I think @jandrieu's interpretation is right.

[peacekeeper]

A DID Method cannot, by my reading, overwrite how the service parameter works.

I agree with this statement. Another way of putting it could be that implementing this feature doesn't require any DID method-specific knowledge, since it operates only on the DID document itself.

[wip-abramson]

@peacekeeper, where is the spec text today though is the made explicit.

Where optional parameters are supported, they are expected to operate uniformly across the DID methods that do support them.

Maybe here.

But DID methods could still overwrite this functionality. It just isn't advised to.

[peacekeeper]

I still don't see any error in the current spec text, but I agree that this could be made a bit more explicit. I think there are three kinds of DID parameters when it comes to how method-specific or -independent they are.

1. DID parameters that work the same way with all DID methods and don't require method-specific implementation, e.g. service, serviceType, relativeRef.
2. DID parameters that work the same way for those DID methods that support them but require method-specific implementation, e.g. versionId, versionTime.
3. DID parameters that are only used for a specific DID method or a small set of DID methods, e.g. oobi for did:keri.

I think it should be easy to improve the current spec text with a simple PR (without major rewrites or fundamental changes).

[swcurran]

Given that the PRs referenced (#311 and #315) are unlikely to be merged, suggest that this be addressed in a targeted PR if any changes are needed.

As a contributor and DID Method creator, I hope the goal of the spec and working group is that like-named parameters operate the same way across all relevant DID Methods, including DID extension parameters.