Click here to monitor SSC

Product Marketing Manager, former technical author, and a helping of content strategy on the side.

Technical communications – the business of eliminating poetry?

Published 25 September 2009 6:31 am

I just got back from the Technical Communication UK conference. It was pretty good, with excellent presentations from (among others) Chris Atherton, Gordon McLean, Matthew Ellison, and Kath Straub.

Looking back over my notes, for Kath Straub’s presentation Reading between the words… Using text formatting to increases persuasiveness and actionability of copy I’ve written:

“O Hai Tech Comms – T. S. Eliot totally pwns yr collective ass”

Oh, that’s going to take some explaining.

So, first off: what was the presentation about, and why was it interesting? The meat of it is that spacing (headings, line breaks, intra-word, etc) can be manipulated for ease of reading and comprehension. One of the phrases Kath used was “virtual commas”, which is to say using spaces as we might pauses in speech, to emphasise semantic blocks.

She gave a couple of interesting pathological examples – cases where something extra is needed to crystallise the meaning of a sentence. One caused an adolescent hissy fit from a couple of the crusty grammarians in the audience, and since I hadn’t realised that inelegantly unclear participles were quite so risqué, I’ll give the simple one instead:

“The chauffeur annoyed the man with the cigar.”

Who has the cigar? You probably can’t fix that with punctuation. You could do it with spoken emphasis, and, Kath argued, with spacing and formatting. You could adjust the intra-word spacing (or even intra-letter) to provide implicit grouping, for instance. You could also break the lines. Break at “annoyed”, and the man has the cigar. Break at “man”, and it’s the chauffeur puffing away instead.

So far, so glib. In practice, you’d re-write the sentence. But this was actually controversial in a room full of technical authors. I’ll get to why, but perhaps anybody with an English degree knows where I’m going with this. To a literary critic, a typography geek, a graphic designer, or more or less anybody who’s read a poem, it isn’t news that mis en page is semantic.

Here’s a nice simple example. Consider the first few lines of The Wasteland:

APRIL is the cruellest month, breeding
Lilacs out of the dead land, mixing
Memory and desire, stirring
Dull roots with spring rain.
Winter kept us warm, covering
Earth in forgetful snow, feeding
A little life with dried tubers.

Practical Criticism 101: The implicit column of gerunds invite reading as a group, and sit not only physically but conceptually off to one side. The commas are doing some of that work, but so is the end position, and the slight pause you get in the flow of sense from the need to correctly find the next line. Kath talked about the cognitive load of “keeping track of who is doing what to whom” as you reach between lines. It’s a bad thing for simple comprehension, but it’s exploitable in a literary context to create layers of faint, potential meanings.

The line splits there create implicit semantic blocks, two to a line, that can be read simultaneously with the flowing sentences. We are invited to consider the single images as well as the continuous sense.

What’s going on here is that potential ambiguities are being carefully marshalled to suggest potential meanings. In Translations, Brian Friel describes the process memorably: “Uncertainty in meaning is incipient poetry”.

This is the opposite of technical communication. From a clarity-focused, information design perspective, it’s as pathological a case as the man with the cigar.

I at least find it hard to deny that the lineation here (or in, say, something like Ed Dorn’s On the Debt my Mother Owed to Sears Roebuck) is a powerful contributor to the way meanings emerge. It affects how we read, how easily we read, which blocks we resolve, and what we associate. It’s semantic information. Ok, so in tech comms, we’re after fewer, simpler meanings, but we want them just as carefully constructed.

So lets poke the bear. How would I get that into DITA without losing content?

For anybody lucky enough to have missed the brouhaha and evangelism, DITA stands for Darwin Information Typing Architecture, and it’s an XML schema that thinks it’s a religion. Ok, that’s a spot unfair, but the fanboys are noisome. DITA is all about breaking down content and separating it into salient reusable blocks, separate from presentation.

Separation of presentation and content is a big old deal in technical communications at the moment, and it more or less makes sense. Lots of people have a valid use case to crank out a chunk of information in multiple contexts and media. These media could well benefit from different, say, weights of header, so why not store the “content” separately?

The thing is, it’s all been taken rather stridently to heart in certain quarters, leading to a knee jerk reaction whenever author-controlled formatting/pagination/lineation is mentioned as anything other than bleak, sulphurous devilry. This is twaddle. We know about things like reading behaviours and the importance of visual affordance. When we mark up a heading, it’s not because having a title is an intrinsic good (although of course, it isn’t unhelpful), it’s at least partly because headings get weighting and users skim for them. Many kinds of presentation are content.

No tech comms standard for markup (that I know of) accommodates salient spacing or mis en page. LaTeX might do some of it, but I’m told it’s fiendish to work with. You could almost certainly have a DITA specialization, and with something like the tools Kath was talking about, could perhaps even automate the tagging, but by and large, this isn’t how we’re currently thinking.

Uncertainty in meaning is anathema to user intelligibility. If we’re going to make sure we’re not writing poetry, there’s definitely value in having poetry’s level of control over semantic blocks.

Of course, it’s fully possible that this is an expensive distraction. I guess the thing to take away is the same as pretty much every other controversial presentation – shut up about X being not your job/not best practice/very modern, and look at delivering the best information you can to the user.

12 Responses to “Technical communications – the business of eliminating poetry?”

  1. Brian Harris says:

    Well said, Roger. The problem with DITA (and any other methodology that strictly separates content from presentation) is that we know from many studies of user behaviour when confronted with online content is that the single most important factor is the visual appearance of the information – the structure and display of the information on the page. Fine-tuning the look of a page can improve its usefulness no end. And this kind of tweaking of the look and feel is exactly what you can’t get from DITA. Each page of content should be, to some extent, lovingly hand-crafted.

    You can’t have it both ways.

  2. Phil Factor says:

    Hear Hear. Until recently, every author had the freedom to get closely involved in the way that his work was presented, even to the point of choosing the paper. Any typographer who has learned his art properly will tell you that even the typeface itself can subtly change the meaning of what is written. To claim that it is right to separate content from presentation flies in the face of hundreds of years of experience. I’ll be the first to admit it is convenient to pretend that text is just text, and can be presented any way I choose, but the truth is that the text, and its precious message is at the mercy of typesetter, designer, editor, bookbinder and Webmaster.

  3. David Farbey says:

    Hi Roger,

    Like many people at Kath Straub’s talk yesterday, I was fascinated by what she said, partly because she seemed to be arguing strongly against some of the dominant current trends in technical publications. What I took away from her talk was that while adding syntactic spacing to texts gave measurable benefits to average and below average readers, as the educational level of the audience went up so the degree of improvement in comprehension went down. To put it bluntly, there was hardly any improvement for people with university level education, because comprehension levels for those people are already pretty high. That may mean that this issue is not such a pressing problem for much of the work we call technical publications.
    Having said that, Kath’s research does indeed raise important questions, and as practitioners in the tech comms community we may need to rethink our attitudes to the doctrine of “separating content and form” in the light of our responsibilities to our readers to provide easily understandable texts. More on my blog at soon.

  4. Roger Hart says:

    Hi David

    You’re quite right there, and I don’t mention it in the blog – her research did show negligible improvements for fluent readers. I guess I’d still be interested in whether we can get information to users faster or more directly by using her techniques to foreground parts of it, but there may indeed be limited value.

    I certainly think it’s oh, naive, blinkered, something like that (but with not intention to be judgemental) to presume we can completely disregard form. Or rather, I think form is more a part of content that we may be accustomed to believing.

  5. Gordon says:

    I wasn’t in Kath’s talk but shame on you David (I think, I may be overreaching here).

    Comprehension levels change, by definition they need to be high to complete a University education but after that they start to diminish as the need is reduced. I have looked and I’m sure I’ve read about this somewhere but … well my Google-fu is FAIL (as the kids would say).

    So if I’ve understood correctly then “That may mean that this issue is not such a pressing problem for much of the work we call technical publications. ” is a perfect example of what you, in your talk called the “Developer Mirror” (aka The Curse of Knowledge (from the Made it Stick book)).

    As people attuned to word choice, sentence structure, grammatical precision then yes I’d imagine most technical writers have a high level of comprehension of written information.

    I’d strongly object to presuming that because my bank of users HAVE a University education that instantly means they have retained that level of comprehension.

    Anyway, I feel a bit cheeky as I wasn’t ever there but this is a bug-bear of mine. Most people just want an answer and would take it scribbled on the back of a napkin rather than wait 10 minutes while someone drafts and edits a sentence to ensure the participles are appropriately clear.

    I’m NOT saying grammar is unimportant as it plays a key part in the presentation of unambiguous information but WE, as a profession, need to remember that not everyone thinks like us (as Justin Collinge amply demonstrated yesterday in his session!).

    Apologies David if I’ve completely got it arse over tit. Wouldn’t be the first time, won’t be the last.

    Anyway, I’m tired and waffling.

    One last thing, Roger, how the hell do I get an RSS feed of your posts? (ohh and the last last thing, why are you making me sign in to leave a comment!! ) :)

  6. Roger Hart says:

    Hi Gordon

    There should be an RSS link in one of the grey speech bubbles on the right (you may have to scroll), but I’ll warn you that I don’t post very often.

    It should now be possible to comment without signing in.

  7. RichardC says:

    Beckett would send the bailiffs round if even a comma of the page furniture was damaged, or flimsiest stage direction was left unflexed. His intolerance seems more a terrible failure of self-confidence, than a belief in the perfect and forbidding statuary of his works.

    On the matter of total authorial control of form, I prefer it when there’s good reason, as in the concrete poetry of Edwin Morgan. Here is one of his seriously immovable poems:

    “Siesta of a Hungarian Snake”:

    s sz sz SZ sz SZ sz ZS zs Zs zs zs z

    Of course the greatest, most beneficent mistreatment of an author’s original intent
    must be Tom Philips’ A Humument, now happily available in an online edition at

  8. Anonymous says:

    Hi Roger,

    I’ve filed my response under A strident defense of mediocre formatting.



  9. PeterG says:

    what about e. e. cummings for heavens sake.

    anyone lived in a pretty how town
    (with up so floating many bells down)
    spring summer autumn winter
    he sang his didn’t he danced his did

    what would “Buffalo Bill’s/ defunct” look like after it has been formatted properly? Long live the space bar as a literary statement.

  10. Anonymous says:

    I agree that a good design is imperative to the convenience and comprehension of information, but what makes a design “good” depends greatly on the audience.

    The benefits of lovingly handcrafting your documentation are irrelevant if your documentation must be 508 compliant or localized.

    Poetry is written by an author for the love of art. Technical documentation is written expressly for use by others and must be accessible to the largest audience possible.

  11. Anonymous says:

    “The chauffeur annoyed the man-with-the-cigar” is the only (and somewhat ugly) punctuation fix I can think of. Should I have scribbled this note on two-layered napkins, or two layered napkins?

    But as you say, rewriting the sentence makes more sense than, in an attempt to prevent a string of words from collapsing into ambiguity, erecting a rickety scaffold of emergency punctuation.

    To get the full meaning of my comments, please print them in Comic Sans on lavender A5 landscape 60gsm acid-free paper.

Leave a Reply

Blog archive