In the section "Specifying a Document's Character Encoding", it is stated that polyglot markup uses UTF-8. It then says that the prefered way to indicate this encoding is with a Byte Order Mark. 

This is not advisable I feel due to: UTF-8 not requiring a BOM [3]; that it could cause problems with applications (apparently MSIE does or did have a problem) and programing languages (apparently inc. Java [4][5]); it causes otherwise valid ASCII to stop being ASCII. 

As such, I would swap the prefered method for indicating UTF inside the document and add a note about using the BOM.

* By using <meta charset="UTF-8"/> (the HTML encoding declaration)(preferred).
* By using the Byte Order Mark (BOM) character (could cause problems in some situations).


References: 
[1] https://en.wikipedia.org/wiki/Byte_order_mark#UTF-8
[2] https://en.wikipedia.org/wiki/UTF-8#Byte_order_mark
[3] http://www.unicode.org/faq/utf_bom.html#bom5
[4] http://bugs.sun.com/view_bug.do?bug_id=6378911
[5] http://bugs.sun.com/view_bug.do?bug_id=4508058