Nowadays, email newsletters are quite popular on the internet. This is especially relevant in the SEO news sphere. While browsing one of the latest newsletters dedicated to eliminating duplicate pages, I noticed that it stated that the canonical tag, or rather the attribute, should be placed on the duplicate page and point to itself. In reality it should not be the case.
How to correctly write and use rel canonical
Let's clarify this ambiguous issue. Why attribute, not tag? Because rel="canonical" is indeed an attribute (part) of a link, not an independent tag. So, here's the correct way to use the rel="canonical" attribute: the canonical link is placed on the duplicate page pointing to the original page. It looks like this: on the duplicate page, which is at https://yoursite.com/duplicate, create an element like:
<link rel="canonical" href="https://yoursite.com/original"/>
Then simply insert this link before the closing head section in the code of all non-canonical versions of the page (duplicates), e.g., https://yoursite.com/duplicate2 or https://yoursite.com/duplicate?id=2. And voila — all duplicates point to the original.
And for dessert — Matt Katts's opinion on rel canonical and its usage: