27 ms·
Has there ever been a breaking change approved by the IETF? I'm not familiar enough with their history to know of precedence. It seems like the spec/standard sh
by hirsin 7y ago
Has there ever been a breaking change approved by the IETF? I'm not familiar enough with their history to know of precedence. It seems like the spec/standard should be approved at least before we go about asking developers to incur hundreds of millions of dollars in updates so their sites continue working in Chrome.
Ironically, this breaks CSRF protection in OIDC authentication systems (except Google, since they don't implement the form_post standard).