Weak requirements language

[privacycg]

Weak requirements language

[j-br0]

websites MAY interpret an expressed Global Privacy Control preference as they find most appropriate

This one might be unavoidable, but this is effectively a meaningless statement. Consider avoiding normative language and instead concentrate on the intended semantics of carrying the signal. The real teeth in this mechanism lies in the legal enforcement part, so explain that more directly rather than use a "In the absence of regulatory, legal, or other requirements" preface to this statement.

This could be addressed by including a cross-reference to § 5 (Legal Effects) after the prefatory phrase.

User agents SHOULD strive to represent what the user agent best believes to be the person's preference for the Global Privacy Control value.

This could easily be a "MUST".

Different jurisdictions are going to have different UX requirements. One country may require an express act to turn on the signal, another may say that the choice of a privacy-specific browser or browsing mode is sufficient to imply intent. Still others may be entirely silent, or they may say that the signal should always be sent by default to accord with the reasonable expectations of most users. I think SHOULD is appropriate to express the general subjective principle of reflecting user intent while affording the flexibility necessary to accommodate varying legal regimes.

[darobin]

Just for context: the MAY there comes from the fact that this spec shouldn't be defining legal requirements. It's not just in the absence of legal requirements: we are explicitly stating that you may do whatever you want, including breaking the law. It's a pretty safe bet that that comes with consequences, but those consequences will hold not because of anything the spec says.

I personally don't have a preference either way on this one; I'm just flagging that this is there because people keep saying "but you can't set law." We don't need to.

The SHOULD is because browser folks tend to be quite fussy about UI requirements. Also, MUST strive to … best believes is really a SHOULD in disguise. (Candidate addition to RFC6919 I guess.) I agree with @j-br0 that SHOULD reflects the fuzziness and leeway that really is there. Again, I'm not married to this specific formulation, but I would be very reluctant to add a MUST that I don't know how to write a test for.