Re: [sensors] What's the rationale of moving security checks outside of normative requirements?

Each condition of the security check used to be a RFC 2119 "MUST" requirement, they're now used as examples for something the user agent "can" do (whatever that means is up to interpretation, and it's arguably non-normative):

> The user agent can define the list of conditions to check if it can expose sensor readings.
>
> The example list of conditions is presented below:

So there's a massive difference between the current state of things and what it used to be. I'm kind of surprised this actually needs to be stated.

-- 
GitHub Notification of comment by tobie
Please view or discuss this issue at https://github.com/w3c/sensors/issues/287#issuecomment-332131947 using your GitHub account