Authoring polyglot documents is harder than authoring either monoglot HTML5 or monoglot XHTML5 documents. In particular, it is a lot harder than authoring monoglot HTML5 documents. Thus, if Web author makes his/her documents polyglot when there's not a real need to do so, the Web author ends up doing useless hard works, which isn't nice.

Considering the history of XHTML advocacy pushing Web authors to try (and most often fail) to make their text/html content XML-compatible without a good reason, I think there's a real risk of history repeating and authors jumping through hoops they don't need to jump through if "Polyglot Markup" is published without better guidance about its applicability.

After all, if the W3C publishes something, readers will assume that the W3C is endorsing whatever is described in the publication without reservation.

To avoid giving the appearance that the W3C advices Web authors in general to write polyglot documents (which in the general case leads to extra work without a return on the investment), please include text in the introduction that communicates the idea that polyglot markup is a solution to abnormal needs and if a Web author hasn't personally concluded that (s)he has use a case that is best addressed by polyglot markup, the Web author is most likely best off using monoglot HTML5 or XHTML5.