Links on delinkification

The airing of the idea of delinkification did not, you’ll be relieved to know, set off a catastrophic implosion of the World Wide Web. It did, however, set a few minds to pondering. At the Columbia Journalism Review, Ryan Chittum considered the distracting qualities of the link and their influence on our ability to pay attention:

Think about reading a newspaper pre-Web that decided it wanted to turn a few words blue here and there. Isn’t that in itself distracting? Now think about how many times you jump in and out of a story to follow some link. It can’t not be distracting.

It’s not a trivial question to ask what the Internet is doing to our attention spans. I know mine, for one, is shot to hell. And my suspicion (totally unproven and based only on personal observation!) is that you scan more than you read on the Internet. With a printed publication, you read more than you do on the Web. Why is that? Well, you’re not distracted by hypertext for one … Dismissing the question of what links do to attention and readability as some anti-link nonsense does nobody any good.

At the Economist, Tom Standage – yes, the chortling twitterer – looked beyond the question of distraction and pointed out that making a piece of writing sufficient onto itself is a worthy pursuit, and a self-contained article or essay can be subtly undermined, rather than enhanced, by inserting lots of links:

I don’t mind piles of links in sidebars, but I find links in text can be irritating if there are too many of them. Of course, it makes sense to link to sources, but links also invite the reader to go away and read something else, and they can imply that the item you are reading can only be understood by reading all the references. At The Economist we do our best to write articles that are self-contained and make sense without the need to refer to other sources, which leads to some characteristic Economist style quirks, such as saying “Ford, a carmaker”. (See? We saved you the trouble of having to ask Google what the company does.) When those articles are published online, there are very rarely hyperlinks in the body of the text.

Writer Russell Davies, on his blog, wondered how the fact that, increasingly, writers don’t know the form in which they’re work will ultimately be read will come to influence how they write. Once you wrote for a book or a magazine or a blog, but now the fate of a piece of writing is often unclear. An article written for a magazine “could end up read on a phone, a tablet, or a kindle. It could be read via RSS or instapaper or something else”:

I know this is terribly obvious and not terribly new. But it’s never fully struck me before. I’m used to thinking the design of things has been atomised, fragmented – that poor old designers for the web could never be sure how something was eventually going to look. But I hadn’t thought about it as something for writers to worry about. It’s just text, how different could it be? But it is different, if you wrote a book you used to have a reasonable idea what the reading experience would look like – no longer. That seems like a thing. That might change writing. A bit. Not a lot. But some.

On the other end of the spectrum, tech blogger and twitterer Mathew Ingram railed against the idea of self-contained pieces of writing. To him, people who don’t pepper their prose with links are lily-livered varmints hiding behind their own words:

not including links … is in many cases a sign of intellectual cowardice. What it says is that the writer is unprepared to have his or her ideas tested by comparing them to anyone else’s, and is hoping that no one will notice. In other cases, it’s a sign of intellectual arrogance — a sign that the writer believes these ideas sprang fully formed from his or her brain, like Athena from Zeus’s forehead, and have no link to anything that another person might have thought or written.

I guess this means that before the link was invented, all writers were arrogant cowards. Which, come to think of it, is probably true.

(Warning: the first idiot who writes a comment on this post pointing out the “irony” of its links will be tracked down, tortured, and shot.)

UPDATE: In a new comment on my earlier delinkification post, Salon’s Laura Miller reports on the reactions to her experiment in arranging links at the end of her articles:

My readers have been overwhelmingly enthusiastic about the change, but Web punditry seem to regard it as a kind of blasphemy. There’s been hysterical exaggeration (pretending that the proposal is to eliminate external links entirely) and other misrepresentations, intentional or not, which surprise me in a community that’s always presented itself as embracing change and flexibility. I wonder if the prospect of overturning a single longstanding tenet of the digital punditry is threatening because it undermines the prophetic powers of Web pundits as a whole (i.e., if they were wrong about this back in 1995, maybe they’re wrong about other things today)? The whole discussion takes on the quality of a doctrinal war.

12 thoughts on “Links on delinkification

  1. Nick Holmes

    (Warning: the first idiot who writes a comment on this post pointing out the “irony” of its links will be tracked down, tortured, and shot.)

    That’s me. But before you shoot me, consider that there’s good linking practice and bad. Why would I be distracted by yours which are so well crafted.

  2. Jason Treit

    The objection about “a newspaper pre-Web that decided it wanted to turn a few words blue” puts stress in the right place: style. The surface features of text. Interface. Not the presence or absence of links. For more down this trail, turn to my other comment.

    Near the end of that same thread, someone added, “Perhaps someone needs to invent an app that makes it impossible to multitask.” Trivial-sounding but again right on point. If findings about hypermedia and diminished attention prove out, interfaces that serve the illusion of a unitask, or a content area with no outbound visual cues, might be what’s next. As long as there’s a hypermedia turbo switch. For more down that trail, turn to Aza Raskin on cognitive shields.

  3. KiltBear

    Part of this, though, is being an educated reader, and the other part is being a good writer.

    Read through first without leaving the page, then go back through and link to interesting things. (I comman-click in Safari which puts the link in a tab behind what I am reading, so it waits for me.) How often do you stop to read each footnote in a book with aggressive citations?

    Write compellingly enough that people won’t want to leave your text.

    Re: app to turn off multiple-tasking: say hello to iPhone OS. Interestingly enough, there are desktop writing applications designed to hide all other functionality of your desktop computer allowing you to concentrate on writing.

  4. Seth Finkelstein

    The neuro-nonsense in the original post put me off (“People who read hypertext comprehend and learn less, studies show, ..” – it’s science!). But the result of rattling the chains of the conference-club was a redeeming factor. That was food for thought (yes, good nutritious whole-grain thought, without any of those junk-food empty-calories unwholesome newfangled stuff, like Twitter/Facebook/Blogs/TV/Radio/Movies/Pulps …).

    A-listers react vociferously against any implication that their business-model (data-mining and attention-mongering for corporate marketing) is not the correct, even moral, way of doing things. In this case, you basically wrote a post that, at heart, said “Lectures by learned individuals are good and productive, and connecting elsewhere during such a lecture is a harmful distraction”. OF COURSE that would send people who base their entire living on monetizing various on-line social interactions, into frothing at the mouth.

    Note it doesn’t necessarily make you correct. It makes the result obvious.

  5. Mikewhitehouse1

    There’s another danger to linking. When I link using a given keyword, I’m sending my visitors to a specific page. That might be OK for a direct citation, but what if there are multiple sources related to a given keyword? Why should that one link I selected be privileged above others? What makes that link the best source? I’m making a value judgment about the quality/relevance of that link, but that’s a highly subjective calculation. I could be steering people in the wrong direction.

    If I wanted to link to a story about BP’s failed top kill, where do I link to? Reuters, AP, NY Times? Are these the best pieces; are they the most factual? It’s a judgment call, and personal bias certainly comes into play.

  6. CS Clark

    I think there is also a social function to links to take into account alongside whether or not they are good for my brain. One thing I consider when using inline links to background information, instead of a specific article, is whether including or excluding them will be more polite. It seems rude to me to assume that people are unable to look for information on Wikipedia themselves, or to make assumptions about what the average reader needs help with, but it also seems rude to me sometimes if I make an obscure reference and don’t back it up with a link.

  7. Alex Tolley

    Links can be somewhat intrusive, but they do serve a solid function. Scholarly articles have always had references, plus marks in the text to indicate which reference was applicable. This is a very useful function. My sense is the standard hyperlink is distracting, whilst more subtle marks in the text would be less distracting to the reading flow, but still ensure the value of links to support the ideas in the text. What this comes down to is a visual design issue, rather than whether links are good or bad per se.

  8. Magpie583

    I am a long time and avid print reader. Hyperlinks (aka Wikepedia type) are annoying, and send you on a crusade to find more information just to finish the first thought you were processing. I much prefer hyperlinks to appear at the foot of the page as I then have a choice as to whether I wish to do further research rather than to get definitions for what I am currently reading. I often will read serial articles online when I want to learn more about a topic, but that occurs mainly when the link is at the foot of the page and invites further reading. I don’t see what the crisis is about putting footnotes at the bottom of the webpage; most of us have read plenty of footnotes at the bottom of pages in print.

  9. Ravi M. Singh

    It amazes me how an idea concerning a simple concept of the link has escalated into what looks like a battle for the existence of the web itself. As much as I like Jeff Jarvis and Jay Rosen, I think they went a bit overboard with charges of trying to “unbuild the web” when placing links at the end of the article as opposed to throughout the piece. At the end of the day, I don’t think it’s a bad thing to leave it to the discretion of the writer. The point is that the link is a powerful and novel tool unique to the web and can be used effectively in different ways. So long as they serve the purpose of providing more information and sourcing for the reader, it really shouldn’t matter where they are.

    I happen to like the links scattered throughout the piece, but on the other hand I understand that they are indeed distracting at times and I know that my attention span has suffered a bit as a result of reading so much on the web. The other issue I sometimes have, and this is probably just me, is that whenever I come across a link I feel like I can’t finish the piece I was reading until I click that link and read the piece to which it links. But this is a matter of disciplining ourselves as readers. I think KiltBear had the right idea – read through and then go through the links, wherever they happen to be. That being said, I happen to prefer my links in context, so I agree with Jarvis and Rosen, but they’re blowing the debate a bit out of proportion.

    That’s just my opinion, what do I really know?

  10. Tom Chandler

    As a writer – and in the interest of getting more done – I’ve largely abandoned word processors in favor of “no distractions” programmer’s editors.

    With that in mind, it’s interesting that my first reaction to “no links” text was a negative one. Are we such children that every shiny bauble distracts us?

    As someone else suggested, perhaps the distraction would lessen if link formatting wasn’t so typographically irritating (it’s a design issue, not a user problem), though I am interested in testing an end-of-article link summary.

    Other questions rear up. Is something lost in the value of the link itself by stripping away its context? Does the link add value to the text beyond a pointer to more information? A deserved emphasis? An added shade of meaning?

    Finally, I’m hardly surprised that a-list bloggers would react badly to the idea.

    Links in posts aren’t offered simply for their informational value; they’re also currency (perhaps an alpha dog rewarding a fawning beta), and while stripping them wouldn’t bring the net to its knees, it would devalue some sites.

    Meet the new boss…

  11. Dave Forbes

    A big challenge with hyperlinks on the whole, regardless of placement, is that they reflect only the author’s context, and most likely not that of the reader, who, by serendipity of a keyword search, happens to stumble across the content. The path the reader creates for themselves, the combination of keywords, is unique and timely, as is the next step they may take. And hard coded hyperlinks cannot enable that kind of personal experience. We have been experimenting with implementing this kind of user experience using a semantic web application, our online demo is at

  12. Bill Wren

    Web theories are nice but for the sake of honesty I have to admit my use of links in a post are more often than not laziness. It involves some work to properly cite a source — it is so much easier to just put in a link and move on. I suspect this is true of a good many people.

    My own preference would be for no links, except perhaps at the end. I’m currently reading “Exuberance” by Kay Redfield Jamison and it is the kind of book I prefer — footnotes and index at the end of the book. No footnotes within the text. I find this less distracting and allows me to simply read, every now and then getting curious enough to go to the back and see what the source might be.

    On the web, I think links are like everything else: one size does not fit all. Some like them this way; some like them that way; some don’t like them at all.

Comments are closed.