Re: [w3c/permissions] Describe the permission store using constraints instead of a full model. (#96)

> +    <p algorithm="relevant-permission-change">
> +      A <dfn>relevant permission change</dfn> for one or more <a>realms</a> |realms|
> +      consists of either the UA receiving new information about the user's
> +      intent or a use of <a>write the permission entry</a> or <a>revoke the
> +      permission entries</a> from a <a>realm</a> whose <a
> +      lt="the Realm's settings object">settings object</a> has the <a>same
> +      origin</a> as one of the |realms|.
> +    </p>
> +
> +    <p class="issue" id="issue-same-domain-permissions">
> +      It's not clear if some current browsers share permissions more widely than
> +      a single origin. For example, when a user grants permission for
> +      https://foo.com/ to use a capability, some browsers may also give access
> +      to any origin with a domain ending in ".foo.com". This specification may
> +      need to allow this.
> +    </p>

When cookies did it, it was a bad idea. I am very much opposed to an endorsement of anything that has a scope larger than scheme and host. I would prefer to retain scheme,  host, and port if possible. 

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/permissions/pull/96/files/6d6282b7405265cabec2c0ee30b69877d77d9dff#r63248089