What was the page number again?

Books use number. On the Web, it is URLs that allow user to reach the right piece of information. And they, too, need to be designed with the user in mind: making sure people understand where they're going, and that where they end up reflects what they were expecting.

An interesting example is articles pagination in blogs. Most of the time, they simply use numbers, with URLs like articles?page=2 or maybe articles/pages/2.

It feels natural. After all that's what books use to identify their pages, and we're used to books. But contrary to a book, the content of a blog changes. New articles appear, and the content that was on page 2 soon gets on the page 3, then 4, 5...

This becomes really problematic for people bookmarking or linking to your site. As you update your site, the content they point to end up not being the one they were originally referring to.

We need to pay a little extra attention to the URL so users can create meaningful links. If they link to page 2, it might be they want to link to the second oldest set of 10 articles. But more likely, they wanted to link because that page had something interesting on it at the time they created the link.

For this purpose, the page 2 is better viewed as the "10 articles published before [date of the first article on the page]". With this logic, the articles listed on this page will most likely always be the same (unless you have a time machine... Can I borrow it ;)). This makes the date of publication a perfect parameter to use in the URLs: articles/published-before/2013-06-06. And as a result, the URL provide more relevant information to the user.

As much as the elements displayed on the page (links, forms...), the URL is something the users will be interacting with. Like in the example, let's try and make them more relevant to what they try to accomplish.