Summary: Writers worry about whether links will distract users. To discuss this concern, we need to begin by distinguishing between imperative links that command the reader to click and passive links that merely make finding ancillary material easier.
Tom Johnson wrote a post recently in which he raised an important question about linking, and referred to an earlier article of mine on the subject. When you refer to another document in a post or article, should you link to it immediately? Tom wrote:
Mark Baker has an in-depth essay on linking, where he explores many of the issues associated with links. I would point you to the post, but whether to insert the link here or elsewhere is at the heart of many discussions about linking. So I’ll briefly delay inserting the link while exploring the risks of doing so.
If I insert the link to Mark’s post here, two potential problems occur:
1. You might immediately jump to his post and, because it is rather long, never return here. You’ll never get my full argument because I’ve distracted you with an inline link.
2. The link to his post may return a not-found error.
The second problem does not concern us here. Things move. People move. It is not a reason not to give anyone an address.
The first problem is the interesting one. Many authors are worried about distracting the reader, and that if they distract them, they may not come back. I have argued before that the writer’s desire to be understood (“get my full argument,” as Tom says) is not really germane to technical communication (or other forms of commercial content). The point is not that the reader should comprehend the text but that they should take the correct (or desired) action. If following a link leads them closer to the correct action, it does not matter that it leads them further away from comprehending the individual text.
Imperative linking puts reader at crossroads
But we should note that links of the kind Tom is discussing (and the link I made above to his post) are not simply distracting. They are imperative. By linking to it in the way I have, I am implicitly telling you to go read Tom’s article (and mine). I am discussing an issue Tom raised in his article, and which was extensively discussed in the comments on that article. In fact, this post is really a response to the comments on Tom’s article. The pretty clear implication of this is that I am telling you to go read Tom’s post if you want to fully understand the points I want to make here, or at least the motivation for this post.
And that, of course, puts you, as a reader, at a crossroads. I, as the author of this post, am asking you to go in two directions at once: go read Tom’s article, and keep on reading mine. You have to choose which you will do first, and whether you have the time and patience to do both.
Imperative references also put reader at crossroads
Now we should note that Tom does not avoid this problem by deciding not to link to my article the first time he mentions it. The implication that you should go read my article (you should, it’s great) is still there whether he provides a link or not. The reference is imperative, even if it is not linked. All that is achieved by not linking is that he makes it slightly harder for the reader to decide to read my article first, rather than continuing in his. Presumably this will result in more readers continuing in his essay rather than stopping to read mine. But the difference is probably small. He tells the reader he will provide a link at the end, so the reader does not have to expend much extra effort, at the point of the imperative reference, to find and read my essay first.
(Tom avoids naming my article, so if he had not provided a link, you would have had a hard time Googling for it. Googling “Mark Baker linking article” would get you lots of hits, since I have written a number of articles on linking, but you would have a hard time telling which one Tom was referring to. As an author, you can obscure the text you are referring to in various ways to try to keep the reader from leaving your text, but then you have to ask yourself, if you are discussing a text, don’t you owe it to the reader to give them access to it? Why should they trust what you say about it if you don’t let them read it?)
Imperative linking and Every Page is Page One
Within the comments on Tom’s article, a couple of people used this example as a basis for criticising the emphasis that I place in rich linking in Every Page is Page One. To answer those concerns, I need to make a distinction. The rich linking that I advocate in Every Page is Page One is not imperative linking, it is passive linking. In fact, I regard imperative linking — telling the reader, implicitly or explicitly, to follow a link — is bad practice in Every Page is Page One.
Distinguishing imperative and passive linking
Here’s the difference between imperative and passive links. In this sentence, I am going to recommend you read a couple of other articles that I have written on linking, such as Why Your Content Needs More Links and The Nature of Hypertext. These are imperative links. I am referring to specific articles and inviting you to read them. Once again I am putting you at a crossroads. In this sentence, I simply want to make the point that the Web is a hypertext medium in which readers are free to forage for information like wild animals. None of these links is inviting you to click on them or to read the articles they lead to. They simply sit there passively.
For a good example of passive links used for this purpose, see this article on quantum computing that contains passive links on its references to quantum mechanics and related subjects: http://www.vox.com/2014/4/9/5588236/quantum-computing-explained. (Yes, this is an imperative link. Choose whether to view the article now or come back to it later.)
Now, some may see an implicit invitation in the mere presence of a link. But when there are six links in a single sentence, it is hard to imagine the reader thinking that the writer is really inviting the reader to click them all. One passive link sitting lonely in a long article might cause the reader to ask whether the author intended the reader to follow it. Consistent rich passive linking (as is ubiquitous on Wikipedia, for instance) does not.
Imperative and passive links have different forms
But mere link density is not the only thing that distinguishes a passive from an imperative link. How they are written matters greatly. A passive link does not change the construction of the sentence that contains it. Nor does that sentence make an explicit reference to the article it references. In fact, you could freely substitute a different resource as the link target for the passive link without changing the meaning or impact of the original article in the least. An imperative link usually, but not always, depends on a specific resource. A passive link never does.
An imperative link in the form “for more information on X see” often does not depend on a specific link target. In many cases, the “for more information on X see” form results from moving a link from its natural position inline to some other point in the text — the end of the article, section, or paragraph. This creates a link in the imperative form where the linking intent was passive. However, the reader cannot necessarily tell that a link in this form has a passive intent. Perhaps the author is saying that you need more information and had better click here to get it. Thus moving the link in an attempt to avoid distraction actually forces the link into a more distracting format.
Purpose of passive linking
But what are passive links for, if not to invite the reader to follow them? They are there because the basic assumptions we make about the reader when we write are wrong. And they are wrong not because we made a mistake in identifying the reader, but because every reader is different: they know different things, they understand different things, they believe different incorrect ideas about different things, and they want different things. No one set of working assumptions can possibly be right for more than a few readers.
We would like to imagine that the best thing a reader can do, even if they are not fully prepared, is to stick with our text to the end. This probably isn’t true. John Carroll pointed out the paradox of sense making, which says that a reader can’t make sense of text that conflicts with their experience and expectation. Sticking with the text therefore isn’t a viable option when the reader does not understand one of the ideas it introduces. But even if it were possible, it is not how readers behave. Readers skip out of content when they encounter things they don’t understand. The role of a passive link is to suggest somewhere that they can go.
Without that suggestion, the reader will probably Google for an answer or post a question in a forum. In either case, they are lost to your content, and to your company. And any inclination they may have had to visit your content next time they have a question has been weakened or destroyed. The link gives you, as the writer, another chance to meet the reader’s needs yourself.
Passive links and the reader’s curiosity
In the comments on Tom’s article, Steve Anderson asks (of links) “If they don’t arouse curiosity, why have them?” This is why: because the subject matter itself arouses curiosity, and, often, frustration. Passive links exist not to arouse curiosity but to satisfy it, not to create a distraction, but to reduce distraction by allaying frustration.
So, in the sentence above with the examples of passive links, the passive links serve to allay the curiosity or frustration of any reader who may not know what a sentence, the Web, hypertext, a medium, information foraging, or wild animals are. Of course, these links are not exactly aimed at the same size of population. The number of people who know what a sentence is is larger than the number who know what hypertext is. The number who know what information foraging is is smaller than the number who know what hypertext is. If you can put your arms around your whole audience and say, they will all know what this means, then there is no point in adding a passive link. If you think some might recognize the term but not fully understand it (hypertext is such a term for my audience) or that many might not recognize the term at all (for my audience, information foraging and paradox of sense making) then a passive link adds value.
Links and non-linear reading
Every Page is Page One is an information design principle that acknowledges the fact that readers, particularly of technical communication, do not read linearly, nor stick to one information source, but forage widely across many information sources. It does not try to keep the reader on the straight and narrow, therefore, but supports the behavior we know they will use anyway by making it easier to move from one source to another. This means both a rich use of passive linking, and writing topics that work as page one no matter where the reader comes from. Its response to any tactic that tries to keep the reader from wandering is: that is a battle you can’t win. Where you can win is by making it easier and more attractive for them to wander into your content and to keep their wandering within your content set.
This is why I think the use of imperative linking is a bad idea in Every Page is Page One. Imperative linking is an attempt by the writer to shape the reader’s course. Every Page is Page One is about supporting the reader finding their own course. Rich passive linking helps the reader to chart a course based on their own needs and interests. Taking the imperative links out is as much part of the Every Page is Page One linking strategy as putting the passive links in.
Imperative links and the discussion of ideas
Imperative links (and imperative references) necessarily play a big role in the discussion of ideas. Very often, a post like this one, or Tom’s post that it references, are explicitly addressing ideas (or the misunderstanding of ideas) that were raised by particular people in particular texts. In many cases, particular people are associated with particular ideas (me with Every Page is Page One, for instance, Carroll with Minimalism, or Horn with Information Mapping), and so any discussion of these ideas tends to reference back to their texts at some point. That is why there is a lot of imperative linking in this blog post, and in this blog in general.
Imperative links in technical communication
Imperative links, on the other hand, have little or no proper role in technical communication. Where they are used, it is usually because the writer is trying to dictate the reader’s course, not because the nature of the material or the user’s need demands it. But technical communication often plunges the reader into a web of unfamiliar terms, tasks, tools, and concepts. And because readers tend to use tech comm opportunistically rather than systematically, avoiding the gentle introduction and diving into the middle, they are often reading material that is full of references to things they don’t know about or understand, each in their own individual way. This is what passive links are for: to enable the reader who has dived into the middle to find a path to the information and the understanding that they need to perform their desired task.