Everyone has different preferences on how they would like to read content. Some prefer lots of links in an article, in order to further explore the subject and learn more detail. Others find links distracting, and would rather read print-like plain text. Even when it comes to citations, people prefer different footnote styles, and there has always been a lot of debate.
In 2010 Nicholas Carr published a piece arguing that authors should delinkify, or more precisely, move hyperlinks to the end of an article. Carr argues that though links are convenient, they also produce a cognitive penalty while reading:
Sometimes, they’re big distractions – we click on a link, then another, then another, and pretty soon we’ve forgotten what we’d started out to do or to read. Other times, they’re tiny distractions, little textual gnats buzzing around your head. Even if you don’t click on a link, your eyes notice it, and your frontal cortex has to fire up a bunch of neurons to decide whether to click or not.
Many people interested in technology and journalism responded, and in fact the Nieman Lab had a series of articles addressing the issue of hyperlinking in mainstream journalism.
Many other news organizations have published guidelines on how to link. The BBC published a set of guidelines it uses to decide how to link and Wikipedia also has an interesting set of guidelines for how to include internal and external links.
Jonathan Stray wrote a piece quantifying how mainstream news websites were currently hyperlinking -- the results were quite different than the rest of the web, in that news articles had an average of 2.6 links per article, and only 1.7 if he excluded internal links to news site topics pages. He contacted various news sites and found reasons why they don't link. First, many organizations use tools that automatically generate links, which easily produce links to topic pages within the site, instead of external links. Second, article content is often required downstream in other tools that don't deal well with HTML content. Third, it could often be confusing to determine exactly what to link to. For instance, many science journal articles are behind a paywall. Finally, there was a pervasive idea amongst news organizations that they did not want to send readers "offsite," perhaps misunderstanding the value and purpose of links.
Jonathan had previously described the purposes of linking as for storytelling, informing, creating currency, and enabling transparency. Though these are all valid reasons for linking, he shows through his own writing how linking can be confusing and cause cognitive overload. He heavily links within this article, and often times he links vague words to articles without providing context in the text about why he is linking or where the user can expect to go. Some links go to youtube videos that would start producing sound immediately. Simply adding links within the text to other stories which cover the same topic can interrupt the reader's flow and overwhelm her before she can even finish the article.
An example that provides how articles can benefit from linking is Joel Spolsky's 2006 blog post where he spent three hours rewriting a post by Steve Gillmor, the original inspiration for Nick Carr's post. Joel added links, context, and definitions to Steve's post which was devoid of links. Not only did fleshing out Steve's post show Joel that Steve's original argument didn't hold up, it also created a much more readable, thoughtful piece. Clearly linking and context has its uses.
Jason Fry, in his Nieman Lab post, brings up three topics to address in order to have a productive discussion about links: credibility, readability, and connectivity. Linking provides a form of credibility in that it shows the author has done research and is willing to back up what they say. The lack of links can be seen as suspicious.
Jason acknowledges that heavy linking does cause cognitive overload and is a problem, but maintains that Carr's idea of putting links at the end of an article won't help because the original context for the link will be lost.
The final point, about connectivity, explains why there were so many negative responses to Carr's post: people were worried about unbuilding the web. Linking less breaks down the fabric of the web.
Though this may be true, Carr wasn't advocating the removal of links altogether, he just wanted them moved out of flow of reading in order to reduce attention loss. This project, Hyperlinker, aims to help authors better use links by providing a set of guidelines and tools to reduce the cognitive overload on the user when reading articles. Using Hyperlinker, a journalist can link appropriately but provide the user the option of reading links in a less obtrusive way.
Hyperlinker provides several tools for the author. First, the author can add fields to link tags in order to specify what kind of link they are making. Based on the link type, Hyperlinker can create additional content on the page to display that information to the user in a helpful way.
Hyperlinker also gives authors a way to link to a set of links using one underlined word or phrase, instead of many. Finally, Hyperlinker creates a button so the user can toggle all the links in the article on or off.