The Reader’s Path Cannot be Made Straight

By | 2014/06/24

The straight path. It is an idea with immense psychological appeal to us. Every valley, Isaiah promises, shall be exalted, and every mountain and hill laid low (Isaiah 40:4). As communicators, we naturally want to lay out a straight path for our readers. But the truth is, we lack the power to make the crooked straight and the rough places plain.

The crooked path and the paradox of sensemaking

A crooked path through a forest.

The reader walks a crooked path through a forest of information.
Simon Carey [CC-BY-SA-2.0 (], via Wikimedia Commons

In his landmark book, The Nurnberg Funnel, John Carroll described what he called the paradox of sensemaking, which can be roughly summed up as saying that learners cannot make sense of the learning materials because they don’t correspond to their current mental model, which can only really be changed by experience. No documentation can ever work perfectly, therefore, because it can only make perfect sense to someone who already understands what it is saying.

In proposing minimalism, Carroll was not saying that it was a perfect system, only that it got in the learner’s way somewhat less than the systematic instruction common at the time. (Unfortunately, Minimalism has become a system in its own right, with the paradox of sensemaking apparently forgotten.)

The paradox of sensemaking tells us that no matter what we do, readers will still thrash, will go off on tangents, and will make mistakes. This is not a failure of instruction, it is how learning works. No matter how attentive a parent you are, your kid cannot learn to walk without falling down and hurting themselves a bunch of times. It is an inherent part of the process. A kid who never falls never runs.

The temptation to straighten the path

Carroll named his book for the legend of the funnel of Nurnberg, which was able to pour knowledge directly into someone’s head. Carroll’s point was, there is no Nurnberg funnel. What I greatly fear these days is that people who refuse to accept that there is no Nurnberg funnel will create documentation sets that are much worse, rather than better, because they are designed to prevent the reader from every straying or stumbling.

Every Page is Page One is about admitting that every reader will take their own course through a combination of content, experimentation, and asking for help, as they work through the process of resetting their mental model. It won’t always seem like the best or straightest course to someone who already has the new mental model – we seem to have an almost limitless capacity to forget how hard it was to learn something once the light bulb has gone off in our own heads.

But that course is driven by the gap between the reader’s current knowledge and the new mental model they need to attain. By making sure that every topic begins by establishing its context, and that it links richly along lines of subject affinity, an Every Page is Page One topic set enables the reader to travel that crooked path more quickly. It does not try to straighten it out, and it recognizes that ever reader’s crooked path may be different. It is an accommodation to the paradox of sensemaking, not an attempt to solve it.

This means we are never going to calibrate the perfect organization or the perfect set of relationships. I believe this is the wrong focus. Our real focus should be on making sure that the individual pages work really well for what they do – something I wrote about in my recent column on TechWhirl:

Even for those with the right model, the path is not always straight

Of course, not every reader is struggling to make sense of the product. Some are experienced users, and some have the right mental model going in. For them, it is simply a matter of needing data, of needing a reference.

Even a procedure can be reference material, if the reader has the right mental model. Carroll found that people with the wrong mental model could not follow procedures, but if you have the right mental model, you may still need to know which order to press the buttons in. The difference between having the right mental model and having the wrong one is that when you have the right model, you have confidence in the instructions and in your interpretation of them.

For people with the right mental model who are simply in need of reference material, the path can be made fairly straight. Perhaps not entirely straight — having the right mental model for the product and the task you are performing with it is not the same thing as having the same mental model of content organization as the author devised. People wanting reference material will still search for it, and, search and search skills being what they are, will often miss slightly, and will need to find a path from where they are to where they need to be. There is randomness in that path too, and we should do all we can to accommodate the random ways in which people stumble the last mile to the reference information they need.

Confidence in the wrong model makes paths particularly crooked

But being confident in your mental model is not the same thing as actually having the correct mental model. While the path is somewhat straighter for the person who actually has the right model, it can be even more crooked for the person who doesn’t but is confident that they do. They will often go barking far up the wrong tree, and waste much invective on the people who designed the product and wrote the documentation, before the light goes off and they realize they have been thinking about it all wrong. They will have walked a long and crooked wrong path by that point and may walk another long and crooked path before they finally get to where they need to be.

Trying to force such people down a single path will not work. It will not help them get to the point of realizing their error, nor help them find their way back. Helping them to wander effectively, on the other hand, may help them reach the point of disillusionment faster, and help them chart a faster course back.

If we cannot create a straight path for the reader, therefore, we should focus on helping them walk their own crooked path most productively. That means two things principally:

1. Make sure that every topic you create does the job it is supposed to do and does it well. If we can’t make the way straight, we can at least make sure the stepping stones don’t wobble. We have no idea of where the reader may come from or where they may be going, so create every page as if it were page one.

2. Provide for rich local navigation. The reader is following a crooked path, so we don’t know what direction they will want to go next. We must therefore provide many options. But even a crooked path is a path. Their next destination is not some random leap. Rather they will follow the scent of the information they need, the same scent that brought them to the current topic. They are interested in subjects that are related in some way to the current topic. They need subject-local navigation, not links to the top of the documentation set.

Want to talk about how to create an information set that helps the reader chart their own course? Get in touch. Let’s chat.

5 thoughts on “The Reader’s Path Cannot be Made Straight

  1. Larry Kunz

    Oh, my. After years of preaching the EPPO gospel, now you’re quoting scripture. 😉 The Isaiah reference was apt – nicely played.

    You’re right about the path being crooked, but I think there’s also something to be said for putting up signposts, or at least scattering breadcrumbs. When I reach the end of the crooked path and the light goes on, I want to turn back and gesture to those who are following: Hey, c’mon. Over here, this is what you’re looking for.

    Perhaps that’s what you had in mind with point #2 at the end. While we can’t neglect the wanders, as you say, some seekers do follow a more predictable path and do benefit from having the signposts — and we can’t neglect them either.

    1. Mark Baker Post author

      Thanks for the comment, Larry.

      I certainly agree about putting up sign posts. But a sign post implies a branch in the path, a decision point for the reader. And that implies the that there are different readers for whom each of the potential paths is the right one.

      Since we can’t know which it the right path for any particular reader — both because they may have different goals, and because they may need to walk different paths in order to prepare themselves to understand the goal when they arrive at it — we have to provide sign posts to help them make their own choices.

      The problem with many topic based conversions is that they break the single linear narrative structured of the original material, with its attempt at creating a single straight path, but fail to put up sign posts to enable the reader to navigate the sea of topics that has been created.

      Sign posts are of two essential varieties. One points to way to a distant place. The other announces that you have arrived a new place. This is why an every page is page one topic needs to both establish its context (the “you have arrived here” sign post) and link richly along lines of subject affinity (the “this way to Hull, that way to Halifax” sign post).

      The reader will wander through your content set because you cannot make the reader’s path straight. How well you put up road signs in your content will largely determine if their wandering is directed and fruitful or pointless and frustrating.

  2. Pingback: Docs that are Part of Larger Systems | Every Page is Page One

  3. Kel Mohror

    Is structured writing (“Information Mapping”) the cause of “What I greatly fear these days is that people who refuse to accept that there is no Nurnberg funnel will create documentation sets that are much worse, rather than better, because they are designed to prevent the reader from every straying or stumbling”?

    Would more extensive cross-referencing / linking mitigate the impact, Mark?

    1. Mark Baker

      Thanks for the comment, Kel.

      Information Mapping is only one form of structured writing. There are many. Some focus on rhetorical structure, some on computable structure, and some on both.

      But I don’t think structured writing contributes much to the idea that we can create a Nurnberg funnel. I think the main cause is what Chip and Dan Heath call “the curse of knowledge” — our inability to grasp what it is like not to understand something once we have come to understand it ourselves.

      One of my pet sayings is, summation is not explanation. At the end of a long period of study we may come up with a succinct statement that perfectly captures what we have learned. That statement seems so perfect to us, so clear and lucid, that we think other people should be able to grasp the same idea from that summary statement alone. (“Summation is not explanation” is an example of such a statement.)

      But it doesn’t work that way. People who have not been on the same intellectual journey and worked their way over the same intellectual hurdles cannot grasp the same concept from your elegant summation.

      And yet, it is those elegant summations that cement the knowledge for the person who has attained it. They are not the icing on the cake, but the capstone that holds the whole edifice together. The temptation to regard them as a Nurnberg funnel is therefore very strong.

      So as writers we end up trying to create the perfect statement that will be the Nurnberg funnel for the reader. But because the perfect statement is only perfect to the person who already understands, it is often far worse at communicating an idea that something that expects that the reader’s road will be rocky.


Leave a Reply