So I thought I knew what "canonical" was for.
As I understood it (and to be fair, most of the examples on the web) it avoids search engines seeing foo.htm
and foo.htm?parameter=1
as duplicate content, by specifying that only one of them is the canonical version.
Therefore it seemed to me to be an issue exclusive to more complex sites (sites with server side processing).
But take a "vanilla" brochure site of say just five pages (Home, AboutUs, ContactUs, OurProducts, Testimonials). That's just five htm files, no server side processing, nothing at all that's going to risk looking like duplicate content.
But a colleague has said we need a rel='canonical' on at least the home page to avoid
http://www.example.com
http://example.com
http://www.example.com/home.htm
looking like duplicate content.
Is he right?