HTML Standard Tracker

Filter

File a bug

SVNBugCommentTime (UTC)
599410202apply wg decision2011-04-12 22:13
@@ -27961,27 +27961,45 @@ interface <dfn>HTMLAudioElement</dfn> : <span>HTMLMediaElement</span> {};</pre>
   type, or it doesn't support the listed codecs.</p>
 
   <p>The <span>MIME type</span>
   "<code>application/octet-stream</code>" with no parameters is never
   <span>a type that the user agent knows it cannot render</span>. User
   agents must treat that type as equivalent to the lack of any
   explicit <span title="Content-Type">Content-Type metadata</span>
   when it is used to label a potential <span>media
   resource</span>.</p>
 
-  <p class="note">In the absence of a <!-- pretty crazy -->
-  specification to the contrary, the <span>MIME type</span>
-  "<code>application/octet-stream</code>" when used <em>with</em>
-  parameters, e.g.
-  "<code>application/octet-stream;codecs=theora</code>", <em>is</em>
-  <span>a type that the user agent knows it cannot render</span>,
-  since that parameter is not defined for that type.</p>
+  <p class="note">
+  <!--END w3c-html-->
+  Only the <span>MIME type</span> <!-- the WG decision started with the next bit, which is not in the style of the spec -->
+  <!--START w3c-html-->
+  "<code>application/octet-stream</code>"
+  <!--END w3c-html-->
+  with no parameters
+  <!--START w3c-html-->
+  is special-cased here; if any parameter appears with it, it
+  <!--END w3c-html-->
+  will <!-- the WG decision that led to this text had a "should", but this is a non-normative section -->
+  <!--START w3c-html--><!--END html--><!--END complete--><!--END epub--><!--END dev-html-->
+  should
+  <!--START html--><!--START complete--><!--START epub--><!--START dev-html-->
+  be treated just like any other <span>MIME type</span>.
+
+  This is a deviation from the rule <!-- in RFC 2046, section 1,
+  paragraph 3 --> that unknown <span>MIME type</span> parameters
+  should be ignored.
+
+  <!-- but not really a "willful violation" since it's not that the
+  types are not being ignored, just that before the type is handled as
+  a type, there's a special case for a particular set of strings -->
+
+  </p>
 
   <dl class="domintro">
 
    <dt><var title="">media</var> . <code title="dom-navigator-canPlayType">canPlayType</code>(<var title="">type</var>)</dt>
 
    <dd>
 
     <p>Returns the empty string (a negative response), "maybe", or
     "probably" based on how confident the user agent is that it can
     play media resources of the given type.</p>

|