Link Spam, and Nofollow
Source: https://www.searchenginejournal.com/how-google-yahoo-askcom-treat-the-no-follow-link-attribute/4801/
Introduced in early 2005, the HTML link relation attribute “nofollow” is a characteristic assigned to some or all of the links on a webpage meant to communicate to a browser that the link is not to be used to generate page rank values. This is important for websites such as Wikipedia or post-filled websites like Reddit which often become authorities (due to the immense number of sites that link into the page), and can be taken advantage of by unscrupulous users who spam the site’s public forums with links to a website of their choosing, which will then bias the assignment of the authority’s authority (points, link juice, etc) to the spammed link. This is obviously undesirable as it allows website designers or users to control the page rank of websites, rather than the usefulness of a website determining its page rank. The “nofollow” attribute solves this issue by allowing website developers on frequently linked sites such as Reddit or Wikipedia to flag a links on their webpage for browsers to ignore while calculating page rank. However, it is worth noting that each browser handles these flags differently, for example, the popular browser Google does not use “nofollow” links to calculate page rank, nor does it follow the link, or index the page the link points to. In contrast, Yahoo! does follow and index the link, though it still does not use the link in its page rank calculations.
Fundamentally, this development in link organization is rooted in a disparity between the mathematical definition of authority, and the human intention behind the concept of a web authority. The mathematical definition of a web authority lends itself to the creation of heavily weighted links from frequently linked to websites regardless of the actual value of this behavior to a user. Nofollow links help to repair this behavior by giving browsers additional information about the links displayed on a page by allowing a developer to mark them as useful (dofollow) or useless (nofollow) links. However, this tool is far from a permanent solution to the issue of discerning the usefulness of a given webpage to a user. Thus, It is likely that browsers will continue to adjust their algorithms through the creation and integration of additional structures like nofollow to work towards better, more accurate search results in spite of the efforts of malicious users seeking to artificially boost a site’s page rank. (for some good examples of browser oriented HTML structuring take a look at www.schema.org)