Are Links A Distraction?

Eric Andersen called my attention to a post by Nick Carr entitled “Experiments in delinkification“, in which Carr argues that links embedded in text are distracting, and that we’re better off treating them like the footnotes they evolved from and putting them in a block at the end of the text. It’s an interesting piece, and I see the merits of his argument. Indeed, I remember trying to read a heavily annotated edition of Nabokov’s Lolita, and it was extremely hard to maintain the flow of reading the novel while turning every few seconds to read about every last entomology reference in the text.

Nonetheless, I feel that links supply context, and I’m a fan of keeping context nearby. Indeed, I find that clicking on a link incurs a much lower cognitive cost than flipping to the back of the book, searching for an endnote. I’ve had readers specifically thank me for including links to Wikipedia entries for technical terms. I assume those readers are fully capable of finding those Wikipedia entries themselves, but that they appreciate the convenience of the links.

Some of the commenters on Carr’s post suggest that we use technology to address this tension between preserving the reader’s focus and supplying nearby context. Specifically, we can use CSS and have a JavaScript button that toggles the link style between visible and invisible. I like the idea of handing readers control of the presentation style, though I still think it’s important to pick a sensible default. At the very least, a document should be self-contained so that a reader can choose if and when to look at the material it cites. The document should also give credit where it’s due, linking to the material it cites in a way that is visible to people and search engines. Beyond that, I think it’s really a matter of author style.

Still, I’m curious what folks here–especially long-time readers–think. Do I link so heavily that it’s distracting? Would it be easier to read my posts if the links were in a block at the end? I write for you, so please let me know how I can make this blog better. I don’t have the resources to conduct cognitive load experiments, but I’m very receptive to comments.

By Daniel Tunkelang

High-Class Consultant.

26 replies on “Are Links A Distraction?”

The question in my mind is do readers complete the post prior to clicking a link and following another trail of thought. That leads to the question of should you open the link in a new tab.


Quite to the contrary, well-chosen links highlight the important technical concepts in the text, making it easier to skim and more fun to read. As long as the links don’t do things on mouseover, I don’t find them distracting, and occasionally useful.


By and large, I agree with Carr and think that many weblog authors overlink their text. Now some weblogs are really just lists of links and in that context it is entirely appropriate.

However, if the weblog take the form of a short article or essay, then links should be used sparingly and only for resources that are essential to the text. For instance, linking “lolita” above is like overkill. Lolita is a well-known novel (It’s the Wikipedia redirect for the term ‘lolita’) and the point of the sentence is clear without the link.

The same argument holds with the links to CSS and Javascript. Would you actually have linked them in this article if linking wasn’t the topic? It seems to me the links add nothing but distration.


Some aphorisms come to mind: “Guns don’t kill people…” “It’s not how many links you have, it’s what you do with them” etc.

In short, it depends.

Links can be a distraction if they are tempting to follow in a way that distracts (sometimes irreversibly) from the narrative.

Links can be useful (as you point out) for documenting some connection.

The manner in which link anchors and link targets are presented can make a difference as well: in addition to in-line links and footnotes, it’s possible to put links in the margins, where they won’t interfere with the visual appearance of the text, but will still appear within the context with which they are associated.

The web browser has trained us to expect link targets either to replace the currently-viewed content, or to open in a separate tab or window. The tabbed approach makes it easier to switch between several pages, but still only lets you see one at a time.

The hypertext literature, on the other hand, is full of examples of other ways of dealing with link targets, including interpolating them into the existing text, displaying them in an adjacent column, in a separate card-like window, etc. Some of these techniques may make the reading experience more pleasurable, more comprehensible, and more usable than the standard web paradigm we’ve grown addicted to.

[No links were followed in the writing of this comment]


Wired had a piece on this effect too.
It made me think about how distracted I typically get at work so I’m going to start doing things like closing email clients for lengthy periods during the day.
A browser plug-in might help with link-hiding


I’d like to reiterate the “it depends” comment from Gene Golovchinsky above but add that it also depends on content. Academic or technical text benefits from annotation in a way that fiction does not. Possibly the same rules of thumb could be used online as already exist in print.


Nick: I appreciate the counterpoint, though I’m not entirely persuaded.

I’m cautious about cultural assumptions, perhaps because I do know people who might associate Lolita with pedophilia but have not read the novel. Also, the link to Lolita doesn’t go to Wikipedia, but rather to the Google Books entry for the annotated version I described. The entomology link does go to Wikipedia, supplying the context that Nabokov was an entomologist (a fact that is probably less known than his being the author of Lolita).

The CSS and JavaScript links–like the cognitive load link near the end of the post–are the Wikipedia convenience links I described in the post. I actually don’t expect all readers here to be familiar with CSS (or with cognitive load), though I concede that the JavaScript link is probably overkill.

I do feel the links provide utility for at least some readers–and that feeling is supported by reader feedback. What I don’t know is how much of a price other readers are paying by being distracted by those links.

It does seem like the key is making sure the links are, as Gregor puts it, well-chosen. But even there I can’t avoid a trade-off, since utility will vary by reader.

Gene: I’m curious if you’ve seen any blogs that use links more creatively to reduce their associated cognitive costs. I’m even more curious if there are existing WordPress plug-ins to support these approaches.


One thing I could do to minimize distraction is to try to cluster links at the beginning (providing context / motivation) and end of the post (suggesting further reading), making the body more of an uninterrupted flow. The only downside I see in such an approach (other than it requiring more discipline from me as a writer) is that it would preclude my liberal use of Wikipedia links for technical terms.


I think that to remove in context links is to try to unring a bell, Daniel. I think that today’s readers, particularly those who read digitally for the most part, and certainly younger ones raised on this style of reading expect the non-linear, contextual style. They may not click out on a first pass, but revisit for details on a second deeper read. Or they may click out at essential moments to clarify a point. In any case, links add value both in additional information, but in lending your argument the weight of research and the social graph.


Something a friend and I have discussed is the idea of using a JavaScript event to allow a reader to flag a link without opening it and essentially using that to let the reader build their own bibliography for the article.

At the end of the article, the reader would encounter a list of the links (with any meta info/description/titling that I wanted) and would have the option of clicking them then to go read or emailing them to an address along with a link and abstract of the article they just read.

If we ever work on implementing it it will most likely be as a WordPress Plugin and/or Drupal extension. The hardest part would be getting authors to take the extra step of providing metadata about the links they use in the item.


I think the distraction is not the presence of the contextual link itself, which I find consistently helpful at this site. The more basic problem across the Web is one of readability of contextual links, which I think are easier to both read and/or scan if presented initially without underlining, so that they read/scan as text and not as a visual speed bump; but with a color that conforms more closely to the surrounding text, but with sufficient contrast to identify the linked text as just that, a link. For example, instead of black text with blue links with underlines, consider black text with bold black no-underline links that change upon hover to blue with underline links. I observe that the presentation of links differ in your posts and your sidebar.

I know Jakob Nielsen would object strenuously to the argument for dropping underlines as an initial state, based on perceived affordance. I would contend that the familiarity and ease with varied visual displays of affordance of links are now so deep and wide among users, as not to be as compelling or as essential as Nielsen suggests.


I don’t find the links distracting at all, actually I find them very useful. And your posts are interesting enough to make the reader come back to them after clicking on those links.


Links belong in web content if only to clarify or amplify on a blog’s content. It’s sensible to link to explanatory material that help to make your point; conversely, it may not make sense to link every organizational name that may appear if a link to a home page is not germane to the discussion.


Michelle: I agree that it’s silly for anyone to crusade in general against in-context links, so I think the more nuanced questions are how liberally and uniformly to sprinkle them in the text. I am tempted by the idea of pushing them to the beginning and end.

JD: interesting idea, and glad to see that Gene’s already prototyped it.

Brian: thanks! And you raise a good point: there may be a better way to present links. I’m open to suggestions, but at the same time I’m wary to confuse readers by violating standard conventions. Though your point about the sidebar being different is well taken, so perhaps I have some latitude for variation.

Raza, Jill, and everyone else: thanks for the feedback! It sounds like I shouldn’t make any drastic changes, but I’ll make more of an effort to avoid gratuitous links and I’ll see if I can concentrate them more at the beginning and end of posts.


I do agree with Jill that over-doing links is a real issue on some sites and thats compounded by people who use “bubbles” that appear when hovering over a link.

Most of the biggest offenders in this area (and I can’t find one right off the top of mny head right now) seem to be people using auto-linking tools.

I have yet to see an algorithm – up to and including Calais that can be trusted to handle links on its own. Many tools exist to suggest links but thats all they should be are suggestions that the author / editor can accept or refuse.

Otherwise you end up with a sentence like:
Google and Microsoft in takeover bid for Yahoo!while Googlealso eyes purchase of Lichtenstein.

That serves no real purpose.


interesting discussion. im currently looking at cognitive load and impact of displayed metadata – so i’ll try to write back after i get a brain scanner… im focusing on when people look at a user interface but dont see the content. maybe its related in this way.


Daniel: I think auto-linking has its place and that place is strictly client side where the browser (person) via browser (app) extensions control what gets linked and where the links take you off to when you click. For example, I am a big fan of Google Finance so if I want a contextual link for Cisco I would rather go to  than say the equivalent page on the Wall Street Journal’s site.

I am a big fan of semantic tagging on the part of authors/editors on the web to help this along, but when a site tries to do too much it gets annoying. I particularly dislike the way the New York Times indiscriminately scripts their site for that “look up” tool that is invoked when you double click on ANYTHING like:

If you need to look up the word TWO you are you really the NYT’s target audience and if the NYT thinks you need to look up the word TWO what does that say about their view of their readership?

My habit of double clicking words just as a sort of tic while reading contributes to that as well. 🙂


There was one nice benefit and unintended consequence of them and other content producers trying new presentation formats… it drove my move towards much heavier use of RSS to the point that I set up a Python based RSS2Email gateway on my office PC in the office so I can just get a short note in my inbox when something new is posted.

Thats how I have been following the comments to your original post. This way I can read it all and just come back to the site to comment. It also lets me monitor for spam comments on all the Information Today, Inc. blogs that I admin.


Comments are closed.