Blog Post

Links: Not Just the Currency of the Web, but the Soul

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Author Nicholas Carr wrote a controversial post recently about the use of hyperlinks in online content, in which he argued that links were a distraction for readers, and were likely to lead to less comprehension rather than more. This idea was an offshoot of Carr’s latest book, The Shallows, which makes the claim that the Internet — and digital media in general — are making society dumber rather than smarter. Now Scott Rosenberg, one of the founders of the online magazine Salon and of a new media-accuracy startup called MediaBugs, has written an admirable series of posts defending the link as the heart and soul of the web. In his original post, Carr described links as “conveniences,” but said they also functioned as a distraction for readers:

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,” he wrote. “Other times, they’re tiny distractions, little textual gnats buzzing around your head.

The author said that research he looked at for his book showed this created a “cognitive load” for readers, and those who read hypertext “comprehend and learn less… than those who read the same material in printed form.” Some prominent writers and media figures agreed with Carr’s take, including — ironically — Laura Miller, a writer and book reviewer with Salon, who argued that links shouldn’t be necessary if writers did their job of synthesizing the topic properly, and said that most people don’t click on links anyway. Carr also got some support from Jason Fry, writing at the Neiman Journalism Lab, and Ryan Chittum in a piece for the Columbia Journalism Review (in the spirit of full disclosure, I wrote about Carr’s argument on my personal blog).

As Rosenberg describes in his first post in response to Carr, much of the research that the author relies on for his attack on hyperlinks and comprehension don’t really fit with his broad thesis. For example, the kinds of links that were studied in the research Carr uses in “The Shallows” had nothing to do with adding context to the text that they were embedded in; in other words, they weren’t the kind of hyperlinks that everyone is used to in blog posts and other Internet content. As Rosenberg notes:

All this study proved was something we already knew: that badly executed hypertext can indeed ruin the process of reading. So, of course, can badly executed narrative structure, or grammar, or punctuation.

Instead of impeding understanding, as Carr and his supporters argue, Rosenberg says he believes that they deepen it, quoting author Steven Johnson as saying that links are a tool for synthesis, “a way of drawing connections between things,” to bring coherence to the vast universe of information online. “The Web’s links don’t make it a vast wasteland or a murky shallows,” Rosenberg says, “they organize and enrich it.” I’m firmly on the Salon founder’s side in this one — without links, what point is there in having hypertext at all? The whole idea behind Tim Berners-Lee’s invention was to enable sites to point to each other and create a “web” of context. Do they impose a cognitive load of some kind on users? Possibly, but in my view, the benefits far outweigh the disadvantages.

In his second post on links, Rosenberg first takes on what he calls “corporate linking,” which is the practice of clogging up text with links “because they provide some tangible business value to the linker: they cookie a user for an affiliate program, or boost a target page’s Google rank, or aim to increase a site’s “stickiness” by getting the reader to click through to another page.” Rosenberg also argues that much of this is Google’s responsibility, because of the value attached to page rank and links:

Google is a great tool because it draws meaning from links. And it is a profitable company because it has placed a tiny but real financial value on many links. But by making links a business, Google also made it harder for editors and writers to defend responsible linking.

In the third post in his series, Rosenberg says that even if Carr is right and links do slow down reading and get in the way of understanding the content they appear in, he would still prefer to have links, because they are “additive and creative.” Links pull together different pieces of a topic and connect them into a whole, he says, and at their best, they also “show a writer’s work” and are “badges of honesty, inviting readers to check that work.” Rosenberg adds that the use of links has multiple benefits, including:

  • Saying hello. “A link to another site can serve as a way of telling that site, ‘I just said something about you.’ This kind of link remains a valid and valuable social gesture.”
  • Showing your work. “Some people are happier with this stuff collected at the end, as we did for centuries in print. But linking in situ gives the reader the information right where it’s needed.”
  • Fairness. “Does a writer present the perspectives of those he disagrees with in a way that they feel is fair? Linking to those perspectives is a way for a writer to say: Go ahead — see if I got you right.”
  • Adding context. “A fragment that gets connected is no longer a fragment. It becomes a working part, a piece of a mosaic, a strand in a web.”

As Rosenberg puts it in the conclusion to his series, writing online without linking “is like making a movie without cutting. Sure, it can be done; there might even be a few situations where it makes sense. But most of the time, it’s just head-scratchingly self-limiting. To choose not to link is to abandon the medium’s most powerful tool — the thing that makes the Web a web.” Hear, hear.

Related content from GigaOM Pro (sub req’d): Why Google Should Fear the Social Web

Post and thumbnail photos courtesy of Flickr users papalars and Robert Brook

9 Responses to “Links: Not Just the Currency of the Web, but the Soul”

  1. So according to Carr SkyNet has already happened, a lot of People are enslaved by their computers :-).

    But links can also be just a simple menu item, allowing me to get here to add my drivel.

    I think what we see here is a dissonance between readers and writers. Both should be able to use links, seems to me many comments are from readers who think links are only for them.

    But links should either get to a “deeper understanding” or “broader understanding” of the article, or as “help” for navigation through a sequence of timed events(menu, history,….). Scientific articles(org. web) are mostly timed events, since they build on each other.

    I avoid context for a reason, it’s basically a real parallel process for the reader and writer. And the “web” doesn’t help much. See Paul’s comment on using parallel tabs.

    Lately we see a lot of articles about how writers/articles can/have/will change. Leaving out readers and their behavior, hence we get into dissonance.

  2. I think links should be used like in text citations. It does not matter how well a paper is written if it is composed completely of citation after citation, you begin to question how much work the author did other than research and compiling data. If you read a web post that is inundated with links in every sentence, it becomes apparent that the author did not add anything of value to the content.

  3. A link may be a great hammer, but not everything is a nail. Granted, the internet (largely) is one big bunch of nails :-).

    I can imagine novelists claiming that links (as evidenced by blogs) are diluting the *mass* that the novel represented. The entire spectrum of ideas, though multi-threaded, was constrained within a single thought-process. Harry Potter never made you look up Gulliver’s travels. Compare that to some of the well-written blogs today – and you see the attention deficit creeping in. Links are here to stay – and resistance is futile, the bloggers seem to say.

    But the internet (and also scientific research, to some extent) is another animal. Picture a butterfly in a garden, flower hopping. Serendipitous discovery (or, stumble-upon, no pun intended) is the essence of the internet.

    Such attention-deficit might not suit the purpose of a novel (or many other similar pursuits). But it eminently suits a large part of internet browsing.

    Having said this, redundancy in links, unrelated promotion, resorting to concocting an article filled with fluff + links and not to mention (certain types of) SEO are some of the serious drawbacks / flaws in link usage. The concept itself is noble, although the same cannot be said of all applications.

  4. personally, without links the web will be a totally boring place with no resources attached. Search engines would do great but not that great. Web resources would be scattered and difficult to get together. To experience the true experience of links, use wikipedia. you would learn 20 things related to the 1 thing you went looking for.

    Students Circle – has launched in the United States and links have been the backbone of our course resources. kindly drop a review

  5. Links are only bad when over-used and not really relevant to the content. But without links, the web would be pretty barren relative to how it is now.

    I’m actually working on a web application ( that will suggest related links as you browse the web, while taking hypertext a step further by crowdsourcing it. It seems to me such a useful idea, to see the most relevant links to any web page with the click of a button.

  6. Here’s another vote for links. Sure, if you’re reading a fascinating entry in Wikipedia, you can click link after link, following a trail so far that you can easily get lost. But that’s what the modern browser is for – opening up a new tab for each step of the read so that you can always trace your path back to the root.

    Links enrich a piece, they show that the writer cares that you might want to investigate further, they are options to explore.

    Long live the mighty link.