Re: [w3c/screen-orientation] Rewrite privacy considerations section (PR #215)

@marcoscaceres commented on this pull request.



> +        A screen's [=current orientation type|type=] and [=current orientation
+        angle|angle=] is a potential fingerprinting vector. To resist
+        fingerprinting (e.g., in private browsing), user agents MAY:
+      </p>
+      <ol>
+        <li>Restrict the value return by the {{ScreenOrientation/type}}
+        attribute to {{OrientationType/"portrait-primary"}} or
+        {{OrientationType/"landscape-secondary"}} to match the screen's
+        aspect ratio.
+        </li>
+        <li>Always return `0` for the value of the {{ScreenOrientation/angle}}
+        attribute.
+        </li>
+        <li>If the screen orientation changes, not fire the <a data-link-for=
+        "ScreenOrientation">change</a> event to reveal a change to a
+        [=secondary=] orientation.

> Wouldn't this encourage battery-expensive polling instead?

Sorry, I might need to word this better. The answer is "no", because the UA wouldn't fire events when switching from "X-primary" to "X-secondary". Only if "X" changes, would the event fire, but it would always report as "X-primary". 

The reason the orientation change event does fire, is that the screen width/height would change, which is already observable either polling `screen`'s attribute or simply by `matchMedia("(orientation: landscape)")`.  



-- 
Reply to this email directly or view it on GitHub:
https://github.com/w3c/screen-orientation/pull/215#discussion_r994074280
You are receiving this because you are subscribed to this thread.

Message ID: <w3c/screen-orientation/pull/215/review/1139966714@github.com>