Closed mikewest closed 3 years ago
@arturjanc @iVanlIsh @annevk: Would y'all mind taking a look at this?
This largely seems okay, except that I don't think this should be the place to start with normative statements around extensions. Encouraging user agents to align along these lines seems okay.
I was trying to walk that line with words like "it would be ideal" and "user agents are encouraged to implement". What struck you as overly prescriptive?
All the RFC 2119 terminology. See the last paragraph of https://infra.spec.whatwg.org/#conformance for replacements. 😊
I've replaced RFC2119 words with non-normative alternatives, and resisted the urge to dive into 6919 too deeply. WDYT?
I took a quick glance at the change and it looks reasonable overall (thanks, @mikewest!)
Ideally, we would be a little more forceful and tell implementers that behaviors such as always sending the Origin
header on extension-initiated requests are important to ensure that sites which use Fetch Metadata headers don't break extensions. But I don't know how to do this non-normatively so ¯\(ツ)/¯
Mostly swiped this text from @iVanlIsh and @arturjanc's conversation in #47.
Closes #47 and #12.