Links are pointers, not connectors. Links are unidirectional by design.
It’s easy to describe the Web as pieces of information connected to other pieces of information through links, but we have to be careful understanding well what “links” really are in this context because it can change completely our perception of the Web.
We use the word link as a shorter version of “hyperlink”. The “hyperlink” term was coined around 1965 by Ted Nelson to describe the mechanism used to connect hypertext documents. The behavior for the hyperlink in the moment of its conception was to connect pieces of information in such a way that the user could explore both ends of the link equally.
The concepts of hypertext and hyperlink inspired many initiatives, and some of them incorporated the terms into their projects, including the World Wide Web. But here comes the important part.
How the web is using the term “hyperlink” is not how it was initially defined. The Web incorporated the link and its apparent meaning, but not the functionality that gave birth to the name of the term itself, world-wide-spreading a misleading concept. Links on the Web behave as pointers, one-way jumps, not real connectors.
You can put a link on your page that points to my page and people visiting your page, because of that link, are provided with an opportunity of visiting my page too. But people visiting my page, can not reach yours from mine if I don’t put a link on my page pointing at your page.