Graphic literacies for a digital age the survival of layout

The Information Society, 28: 236–252, 2012Copyright c© Taylor & Francis Group, LLCISSN: 0197-2243 print / 1087-6537 onlineDOI: 10.1080/01972243.2012.689609

PERSPECTIVE

Graphic Literacies for a Digital Age:The Survival of Layout

Robert WallerThe Simplification Centre, London, United Kingdom

Page layout is dominant in many genres of physical documents,but it is frequently overlooked in academic analyses of texts andin digitized versions. Its presence is largely determined by avail-able technologies and skills: If no provision is made for creating,preserving, or describing layout, then it tends not to be created,preserved or described. However, I argue, the significance and util-ity of layout for readers is such that it will survive or reemerge.I review how layout has been treated in the literature of graphicdesign and linguistics, and consider its role as a memory tool. Idistinguish between fixed, flowed, fluid, and fugitive layouts, deter-mined not only by authorial intent, but by technical constraints.Finally, I describe graphic literacy as a component of functional lit-eracy and suggest that corresponding graphic literacies are needednot only by readers, but by creators of documents and by the infor-mation management technologies that produce, deliver, and storethem.

Keywords graphic design, layout, literacy, multimodality

This paper is about page layout—the juxataposition oftext elements (for example, chunks of prose, illustrations,and headlines) in multi-column pages. It is an aspect of textthat has not been as widely studied as some others and thatis easy to overlook. It is largely (but not entirely) absentfrom literature, and it is therefore correspondingly absentfrom literary studies of text, and absent from theories oftext that originate in literary studies. Instead, it is asso-

Received 17 October 2011; accepted 7 March 2012.Thanks to Leopoldina Fortunati, Judy Delin, and Martin Evans for

their helpful comments. Leopoldina also suggested the useful conceptof layout as infrastructure for reading and writing.

Address correspondence to Robert Waller, Simplification Centre,CAN Mezzanine, 49-51 East Road, London N1 6AH, United Kingdom.E-mail: [email protected]

ciated with nonliterary genres such as children’s books,user manuals, catalogues, and newspapers, which havebeen less intensely studied. Because layout is a nonlinear,holistic quality of text, it is hard to define and to quantify.And it tends to get forgotten when new technologies forcreating, storing, and retrieving text are developed.

In this paper I want to position layout as a less periph-eral feature of text than it has often been considered—asan important infrastructure for reading and writing in anage when few make time to engage with long linear texts.I demonstrate its function and the key role that technologyhas played in enabling or suppressing layout in differenteras. I then discuss its role in enabling effective reading,and I review some of the theoretical approaches that havebeen proposed within different disciplines. Then, look-ing forward, I discuss future types of digital text, arguingfor the continuing requirement for graphic layout even indigital text, and for its incorporation in our definition ofcommunication competence and literacy.

Traditionally, readers have almost always encounteredtext in the context of a document: an object with borders,with a declared aim, with a defined authorship, and withina recognized genre—with all the conventions, rules, au-thority, and audience expectations that are implied by that.Text (language string) has usually been situated in a text(document).

In the new digital culture, though, text is frequentlyencountered as search results—fragments detached fromtheir document context, alongside fragments of the vastunmediated mass conversation that is social media. Andsimilarly when scholars discuss language and informationas an abstract concept, it is often in the form of stringsof text that we can easily store (for example, in languagecorpora), search, and analyze, rather than contexualizedin physical documents.

There is general agreement that through technologicalchange we are experiencing a very major shift in the way

236

GRAPHIC LITERACIES FOR A DIGITAL AGE 237

we communicate, and what we communicate—perhapsequal to the development of printing. There are well-known arguments1 that the way in which we frameour thoughts for the communication and preservation ofknowledge has a profound effect on how we think. In ourown era we face the loss of authority, as knowledge be-comes crowd-sourced through social networks, and withthe loss of coherence as we experience information infragments via a search engine. Both of these are reasonswhy layout might be going out of style—representing, asit does, a carefully considered, editorially mediated, anddesigner-crafted presentation of a complete message.

Paradigm shifts (and some would claim that is what isrepresented by digital communications) are traditionallycarried out to the sound of exaggerated debate between oldand new, which can be difficult for those who appreciateaspects of both positions. In Plato’s Phaedrus, Socratesfamously remarks to Phaedrus that writing would “cre-ate forgetfulness.” He was right—we no longer memorizevery much by rote, but we use writing as a memory tool in-stead. Like any physical tool it extends our reach, gives usfocused functionality, and multiplies our strength throughleverage.

Experience shows that new communication technolo-gies rarely render old ones completely extinct. Instead,they more typically occupy a new niche that was not pre-viously possible, with the old technologies often survivingbut in a less dominant position than before. Theatre, cin-ema, TV, DVD, and YouTube happily coexist. While thedevices that deliver them might converge, it appears thatthere will still be distinct audiences and occasions for liveperformances as well as stored ones, shared experiences aswell as private ones. People still memorize the alphabet,mathematical tables, songs, poems, and speeches. Andeven parchment scrolls still exist for ceremonial functionssuch as academic degree ceremonies, in which a documentplays an important performative role.

That having been said, there is an inevitable momentduring a communications revolution when a traditionalchannel or technique seems to be doomed to extinction,and this is a good time to assess its usefulness. Pagelayout is a little-discussed aspect of text, but it con-nects closely to a range of fundamental issues concern-ing the nature of text, documents, writing and reading.Do we still need it, what are at the apparent threats to it,and are there any reasons to suppose it might survive orreemerge?

A book reviewer of this area remarked that

one problem besetting the theorization of multimodal dis-course is that most senior scholars entering the field havebeen monomodally educated: they are linguists, or musicol-ogists, or art historians. Inevitably, they are thereby biasedby their original field of study, and limited by their restrictedknowledge of other disciplines. (Forceville 2007, 1236)

To declare my own focus and bias: My field of study(and practice) is typography and graphic communication,a relatively immature field largely focused on professionalpractice rather than theory.

WHAT LAYOUT DOES

Let me first use a practical example to establish what Imean by layout, and to explore what layout adds to a text,and what is lost if it is absent.

Consider this double-page spread from The Guardiannewspaper (Figure 1). The long dark bar at the top groupsthe whole spread under its wing. It says this is all one story.The bar is dark red in the original—it is worth noting thatonce we reproduce an actual, material text in an academicjournal we lose key aspects of its reality: color, size, depthand texture. And feel, sound, and smell—often remarkedon by readers of paper documents.2

The larger heading, on the left, dominates the spread,and, reinforced by the dominant image, sets up themetaphor that defines the editorial direction for the spread.The image communicates on an emotional level—both insetting up the atmosphere of discomfort, and in reducingthe word count on the page to make it more inviting forreaders who might be daunted by two pages of solid text.

The charts (bottom left) and glossary (column 2) add anauthoritative tone that backs a view that might otherwisebe seen as just editorial comment (signaled by the authorbeing identified not only by name but with a photo). Theevidence about the human impact of a stagnant economyis supplied by the case study in the center (on a beigebackground in the original), and a supporting voice issupplied on the right by a column headed “Analysis.” Allthis is visually signaled by distinct graphic zones, andhieriarchies of differently sized headings that supply thediscourse cohesion that in a purely linear text would besignaled in words.

One way to judge the function of layout in a spreadlike this would be to remove it—to turn it into a singlelinear text, with no differentiation between sections (acommutation test, in effect3). That is exactly what thenewspaper does on its own website (Figure 2).

The same story in its Web version is stripped of itslayout. The connected stories (the case study and the anal-ysis column) have disappeared from view—they do existelsewhere on the website, but there are no direct links tothem. And although the paper version includes just oneunrelated element (the advertisement), the Web page hasnumerous links to unrelated stories and sections that seekto distract or divert the reader.

The reader of the paper version can slip easily betweenrelated stories because cohesion within the set is providedgraphically: their physical location, the typographic hier-archy, and visual genre distinctions all provide cohesion

238 R. WALLER

Left. FIG. 1. A spread from The Guardian, 26 January 2011. In the colour original, thepanel below the photo, and the glossary, have a tinted background. Original size 470 ×315 mm. From Guardian News & Media Ltd (2011).

Right. FIG. 2. Composite screen image from The Guardian website, accessed January26, 2011. It may not be clear from this reduced-size reproduction, but the links in themiddle and right columns do not relate to the story on the left. From Guardian News &Media Ltd (2011).

cues that in the Web version are absent or are entirely lex-ical. Importantly, the related stories are physically parkedon the same page as the story being read at any one time.This means they are hard to lose track of, and rememberingtheir presence adds little to the reader’s cognitive load.

LAYOUT AND TECHNICAL AFFORDANCE

Although the Internet is usually assumed to be the moreinteractive experience, the reader of the online version ac-tually has the more linear experience at the page level,although readers can still look back and ahead within thestory—and, of course, they have the huge benefit of be-ing able to search electronically, and connect directly tointertextual references or citations.

E-readers are very well accepted by readers of fictionand are overtaking paper books in sales. But they areless well accepted by people who need to study. Usersof e-readers or smartphones have a restricted view, andevidence is appearing that suggests that currently availabledevices are struggling for acceptance by readers whosetasks are not simple and linear (see review by Thayer et al.2011).

Attempts to introduce e-readers for academic studyconsistently disappoint:

Students often mark up texts, seek out and assess references,multitask while reading, and generally do more than just readthe words on the page or screen. Similarly, academic workinvolves a variety of navigation techniques, such as cross-referencing information within a text and across multipletexts. Studies of e-readers in academic environments indicatethey are imperfect devices for these activities. (Thayer et al.2011, 2918)

It is not entirely surprising that e-documents of thiskind seem to have missed the mark for readers—it hashappened before. When new technologies are developedfor text, when we discuss it and analyze it, when we designsystems to store and retrieve, there is a consistent defaultassumption that text is little more than a linear string ofwords and sentences.


Viewing medieval manuscript books one is struck bytheir typically close integration of the visual and the ver-bal. I could cite any number of examples and collections,but the Schoenberg Collection (as represented in Black2006) is particularly striking as it consists largely of in-formation documents, rather than religious or philosophi-cal texts. But when Gutenberg developed moveable type,a side effect of his communication revolution was thatthe typically high integration of image and text found inmanuscript books was largely lost, only reappearing on alarge scale with the invention of chromolithography andphoto-engraving, and with the growth of mass literacy,newspapers and magazines. Printing had such an impacton the spread of knowledge, science, and education that therelative poverty of its graphic form could be overlooked.

Something similar is happening, possibly temporarily,with the development of electronic publishing. When hy-pertext emerged in the 1980s it was heralded as a releasefrom the pure linearity of text, as if we were still in the daysof cuneiform, with no headings, contents lists, indexes ordiagrams—for example:

Unlike the static form of the book, a hypertext can becomposed, and read, non-sequentially. (Landow & Delany1991, 3)

Text is typically presented in linear form, in which thereis a single way to progress through the text, starting at thebeginning and reading to the end. (Foltz 1996, 109)

These quotes (which I could have selected from any num-ber of papers on hypertext in the 1980s and 1990s) per-fectly illustrate the linearity assumption—that text, likespeech, is linear by default; that it is produced in a linearway, and that it can only be consumed in a linear way.

Figures 3–6 show books relating to the cultivation ofplants from circa 1100, 1929, 1965, and 2010. They typifythe age of manuscript, of letterpress, of offset lithography,and of the first generation of digital books. Page layoutbreaks out when freed from linear production technolo-gies, but is suppressed when the next technical develop-ment reverts to the linear default.

Adobe Systems Incorporated, maker of page layoutsoftware, is unapologetic about the lack of support forlayout in its eBook format, in this note from its supportpage:

“Why does my eBook look different than my InDesign doc-ument? The EPUB format does not define page structure,so all the content flows together in one continuous linearstream. This can present a problem for publications that havean elaborate design. For example, if your InDesign documentcontains a lot of sidebars and images that are surrounded bytext, they are linearized in the eBook, so it will look quitedifferent than the original layout. However, if your layout isquite simple, you probably won’t notice much of a differencebetween it and its eBook equivalent. (Adobe Systems, Inc.2010, 2)

This rather retrograde assumption of a linear norm willprobably be temporary, and some newer formats an-nounced for online textbooks have paid rather more atten-tion to the needs of readers and the demands of complexcontent.4

LAYOUT AND READERS

Strategic Reading

Why are some kinds of e-documents accepted by read-ers (novels) and others (textbooks) less well liked? Whenwe read a novel we engage in a style of reading that issometimes called “receptive,” “linear,” or “close” reading.Unless we are reading it as a student or critic, we followthe narrative at a fairly even pace, controlled by the writer.

Studying, in contrast, is an example of what we mightcall a selective or strategic reading process (Paris & Myers1981; Pugh 1975). Strategic readers use a document,or a set of documents, to achieve a goal. They engagein receptive reading for some of the time, but monitortheir understanding, and their progress toward the goal,in a process known as metacognition (Brown 1980) orexecutive control (Britton & Glynn 1987). They thenadjust their style of reading in response to this internalmetacognitive feedback.

Strategic reading is enabled by what the typographerBeatrice Warde5 called the “three great privileges of print-ing” (to turn back, to look forward, and to stop and think),and is echoed in Daniel Pennac’s (2006) classic The Rightsof the Reader (his 10 rights include “the right not to read,”and “the right to skip”).

Pugh (1975) identified five strategies: Receptive read-ing is reading from beginning to end with little variationin pace—appropriate for the novel reader, and well suitedto e-readers, but less appropriate to intensive study orproblem solving; responsive reading means an active en-gagement with the arguments in the text, with frequentchanges in pace, pauses, and rereading; skimming is aquick read to overview the structure or content of a text,either before or after a full responsive read; searchingmeans looking in a general way for answers to a ques-tion; and scanning means searching for a specific word orphrase. Of these five strategies, electronic documents areparticularly well suited to receptive reading, skimming,and scanning (using the search facility in a browser ore-reader).

Studying is not the only strategic reading activity. In-formation documents of all kinds need to be read strategi-cally. No sensible person chooses from a catalogue, setsup a DVD player, selects a hotel from a travel guide, orlooks up a word in a dictionary by starting on page 1 andreading through until the end.

240 R. WALLER

FIG. 3. An 11th-century herbal: Ps. Apuleius, Herbal Eng-land, St. Augustine’s Abbey, Canterbury. 11th century, circa1070–1100. Ashmole 1431, fol. 018v-019r. Copyright BodleianLibrary, University of Oxford. Reproduced with permission.

FIG. 5. Page from The Rose Expert (Hessayon 1967).

FIG. 4. Spread from Everyday in My Garden(Farthing 1929).

FIG. 6. Kindle edition of Allotment Gardening (Berger 2005;Kindle edition 2010). Note that page breaks have no regard tocontent, so widows (isolated words and lines at the top of pages)are common.


Strategic reading is at the heart of document literacy.Along with prose literacy (which measures the fluencyof receptive reading) and quantitative literacy (the basicarithmetic needed for everyday life), document literacy isa key aspect of a wider term, “functional literacy” (Or-ganization for Economic Cooperation and Development[OECD] 1997). It refers to the ability to use documentsto achieve purposes and solve problems. The tests used tomeasure it mostly use visually organized documents ratherthan continuous prose: for example, forms, timetables, in-structions, and user guides (Evetts and Gauthier [2005]include many examples). In many schools and in adultliteracy classes, readers are taught active reading strate-gies: the use of access structures such as contents listsand headings, the use of multiple sources, and ways toapproach different document genres. In fact, the conceptof literacy has become increasingly broad, and the Inter-national Reading Association now includes visual literacyin its definition (Edwards 2010).

Of course, the deployment of these skills dependson documents that afford, allow, and encourage thesestrategies: providing readers with what they need to readstrategically. Textbook designers know that students needinformation to be broken into chunks, and well suppliedwith headings, illustrations, notes, and meta-level studyaids.6 Together, these comprise what Anderson and Arm-bruster (1985) call a “considerate text”—a term that mightbe more widely applied to any text that reflects the needsof readers, rather than just expressing the topic structuresand arguments of the writer.

Linearization in Language

At its most basic level, language is obviously linear—inmost languages, word order is critical within thesentence—and explaining the sequential, or syntagmatic,relationship between words has been a major preoccu-pation of linguistics. Together with the principle of theprimacy of speech, this explains why until fairly recentlylinguistic scientists rarely acknowledged graphic aspectsof text.

Above the sentence—at the level of paragraphs, sec-tions, chapters, or stories—we become increasingly lessdependent on syntax and more on the presence of explicitstructural or cohesive cues. As a simple example, the ex-pression “on the one hand” in English tells us that oneaspect of an argument is about be presented, and thencontrasted with another, which will be announced by “onthe other hand.” In effect, a diagram is being constructedverbally. “On the one hand” is an example of a cataphoric(looking ahead) reference, which requires readers to cre-ate a mental representation that is referenced by somethingthey later read. “On the other hand” is an anaphoric (look-ing back) reference, requiring readers to consult a mental

representation of the text they have already processed.In other words, it requires to them answer the impliedquestion “other than what?” from their memory of thepreceding text.7

It is a small step for designers to see those kinds ofstructural cues as opportunities to turn that putative mentalrepresentation into a diagrammatic representation8—forexample, through bulleted lists, diagrams, numbered steps,or marginal panels. These are instances of documents asmemory tools, reducing the need for readers to constructand refer to mental representations of content structure.

Layout for Strategic Reading: Overcoming theLinearity of Language

Our writing system has also evolved in support of theserhetorical structures: Documents in their modern formwork as tools for overcoming the linearity of language.Although this is often cited as a unique advantage of dig-ital channels, it is not new. The history of paper doc-uments shows the development of an increasingly richrange of ways to overcome the linearity of language and tomake written information accessible: word spacing, punc-tuation, the codex, headings, page numbers, typographicstructures, indexes, and multimodal layouts evolved overcenturies. They moved the act of reading from a sloworal process to the fast, silent, and strategic process thatwe have discussed, in which effective readers deploy arange of strategies to achieve their goals: searching, skim-ming, recapitulating, and note-taking, as well as linearclose reading.

It is not completely clear when silent reading developed,but Saenger (1982) argues that it was the development ofword spacing in the early middle ages (seventh and eighthcenturies) that freed readers from the linearity of slow oralreading, and transformed the way we study, and by thelate Middle Ages, this had developed into a sophisticatedsystem for the visual organization of text.

The complex structure of the written page of a fourteenth-century scholastic text presupposed a reader who read onlywith his eyes, going swiftly from objection to response, fromtable of contents to the text, from diagram to text, and fromthe text to the gloss and its corrections. (Saenger 1982, 393)

This is how designers intend us to read modern structuredbooks, too, although there is only a limited published lit-erature accounting for their layout principles.

THEORIZING LAYOUT

Layout in the Literature of Graphic Design

We can find several alternative approaches to layout inthe graphic design literature. This is typically presented inthe form of manifestos, textbooks, portfolio collections,

242 R. WALLER

and memoirs—not peer reviewed in the academic sense,although often intelligently curated by an editor or pub-lisher. And of course most designers speak through theirwork, rarely stopping to articulate what they are doingexcept to teach students or coach employees in a studiosetting. Schriver (1996) provides a good account of thegraphic design literature, linked to related traditions intechnical writing and usability research.

First, designers use perceptual principles establishedby the Gestalt psychologists (Wertheimer 1938) to accountfor graphic relationships among elements of a page. Forexample, whatever their actual content, we tend to assumethat things that are physically close on the page are relatedin some way (the proximity principle), and that thingsthat look similar are members of the same category (thesimilarity principle). Although no longer current amongpsychologists, for designers these principles usefully com-prise what might loosely be called a visual syntax of thepage, and they are widely used in design education.

There is also a strong tendency among design textbookwriters to focus on formal or aesthetic qualities such asrhythm, contrast, tension or balance. The use of visualform to direct readers’ attention was well articulatedby designers from the Bauhaus and New Typographytraditions, influenced by the Gestalt psychologists, aswell as art movements such as de Stijl and Constructivism(see Kinross [1992] for a good history and analysis). Inparticular, the Czech designer Ladislav Sutnar, a pioneerof designing in double-page spreads, articulated anddemonstrated coherent theories about what he calledfunction, form, and flow (Heller 1993; Sutnar 1961;Sutnar & Lonberg-Holm 1944). However, it still remainsfor such ideas to be integrated into a broader functionalaccount of layout as a component in discourse, and asan infrastructure for writing and reading. In the handsof many design textbook writers these formal graphicqualities are treated as the counterpart to poetic andexpressive qualities in verbal language, so they relatemore to reader engagement than comprehension.

Twyman (1979) demonstrated the wide range of graphiclanguage that can coexist within a single taxonomy thatdistinguishes between the mode of symbolization (prose,numbers, pictures, schematics) and the mode of configu-ration (how elements are ordered and accessed in linear,semilinear, and nonlinear ways). This tradition of thoughtled in turn to the idea that typographic pages are diagram-matic, extending the function of punctuation within the lin-ear text to the page level, displaying relationships such assegmentation, sequence, balance, and salience graphicallyrather than lexically and syntactically (Waller 1982; 1987).This work was an attempt to account for the illustrated ref-erence books that emerged in what we can now see as agolden age of layout, the 1970s and 1980s. Publishers suchas Time-Life, Reader’s Digest, Dorling Kindersley, and

others developed a new genre that, inspired by magazinedesign,9 used the double-page spread as a unit of mean-ing. The diagrammatic quality of these books—typicallyon hobbies, sports, history, or travel—brought layout tothe fore. They were developed by multidisciplinary teamsin much the same way as films are produced: Unlike thetraditional book, in which the author’s voice is primary, inthese books, the writer fills in spaces to order, and providesfunctional text such as descriptions and captions on requestfrom editors, illustrators, photographers, and designers.

Then there is a generic perspective used in design.Layout is the main signifying feature of many familiardocument genres: for example, newspapers, magazines,textbooks, user guides, packaging, and reference books.These everyday genres owe their very being to their lay-out. When readers see them, they know what they are, andwhat to do with them. The graphic layout of such genreseffectively contains the rules or affordances for their use:Engaging layouts and large headings invite the magazinereader to browse; the orderly layout of a user guide in-vites systematic reading, referencing a task outside of thetext through diagrams, and providing large numerals as avisual target to the returning reader.

Considered as “rules for use,” such aspects of layoutcan be thought of as access structures. In earlier work,I have distinguished this from the complementary use oflayout to convey topic structures (Waller 1991), whichare motivated by structures inherent in the author’s topicas distinct from the reader’s task. Together with artifactstructure (which arises as a by-product of manufactureand is unmotivated by communication goals), characteris-tic combinations of these structures account for the typicalstructure of document genres of the kinds just listed. Delin,Bateman, and Allen (2002; see also Bateman 2008) fur-ther developed this model into a fuller account of genrestructures, elaborating in particular the notions of artifactstructure and topic structure.

Genres are natural categories identified by a languagecommunity—the primary evidence for their existence isthe development of a name: magazine, newspaper, text-book. When we need a new name we invent one (“blurb,”for example, to describe publishers’ eulogies), and some-times we subdivide genres into new subgenres (newspa-pers became broadsheets or tabloids).

A related concept is pattern language, with the keydifference being that this is an exercise in naming com-mon configurations that exist but, unlike genres, have nonaturally developed name. It originated with the architectChristopher Alexander (1977), who developed names forsuccessful configurations in towns and buildings. His pat-tern descriptions include definitions of common problemstogether with recommended solutions. They not only offerarchitects a repertoire of solutions that, Alexander argues,reflect the way human settlements naturally evolve, but


they provide names that enable the patterns to be dis-cussed and specified. This approach was taken up in asignificant way by software engineers who needed to finda way to describe common programming objects (Gammaet al. 1994), and from there it was picked up by interfacedesigners (Tidwell 1997/2005) and eventually by docu-ment designers (Waller & Delin 2010; Farkas, Larson, &Naranjo 2011).

These approaches (formal, diagrammatic, genres, andpatterns) have one thing in common: They assume thatreaders combine a focal awareness of the words they arereading or the part of a picture they are inspecting, with asubsidiary awareness of the whole graphic page. “Focal”and “subsidiary” are the terms the philosopher of scienceMichael Polanyi (1969) used to describe a form of holisticperception he called “physiognomic.” At one level readinga page is a little like recognizing a face—you don’t inspectthe eyes, the nose, and the mouth separately but in one take.This makes layout challenging for technologies or analyt-ical frameworks that fail to go beyond the linear default.

Linguistics, Semiotics and Layout

Given that for the most part there is little authorial con-trol of layout, and therefore little intentionality imputedto it (in a novel, for example), it is not surprising that formany years it was largely ignored within linguistics. Butduring the 1990s there began to be growing interest in mul-timodality, and in the extension of discourse analysis tographic aspects of documents. This work typically drew onstructural linguistics, discourse analysis, and genre theoryas its starting points. In an important early contribution,Bernhardt (1985) ranged genres on a spectrum from thevisually informative (in which layout and typography vari-ation are prominent) to the visually uninformative (lineartext), and explored the parallels between the two in termsof Halliday and Hasan’s (1976) structural linguistics.

Kress and van Leeuwen (1996) similarly usedHalliday’s structural linguistics as a starting point butwithin a social semiotics perspective. They develop aninfluential account of the “grammar of visual design” thatthey apply to a range of multiple modalities. Addressinglayout, they suggest that verbal and visual elementsinteract in three key ways to create cohesive pages.Salience refers to the manipulations of the viewer’sattention through such things as relative size and contrast.Framing refers to the dividing off or enclosing of textelements. So far so good, but their concept of informationvalue is more controversial. It asserts that differentzones of a page—left, right, top, bottom, center, andmargin—carry distinct significances. For example, theyassert that the left-hand side of a multicolumn page carriesgiven information (things the reader already knows) andthe right-hand side carries new information (extensions to

the given, things at issue). Although they produce someexamples of, in their terms, grammatical or ungram-matical pages, these are quite limited in number, seemto be selected to demonstrate the point, and underplaythe variety of layouts that can be found. Moreover, theleft–right distinction appears to assume we read pages ina linear manner, whereas the way information is, in theirterms, framed or given relative salience may encouragealternative sequences. Consider the examples I introducedearlier. In Figure 5, the left-hand page shows how to plantthree kinds of rose. Each is given a similar heading, and theorder is almost certainly dictated by the proportions of thepictures: Standards and climbers are both tall and narrow,so fit well side by side, leaving bushes and the descriptivetext to fit in with each other. I cannot see Kress andvan Leeuwen’s semiotic significances instantiated in thispage, beyond the uncontroversial notions of framing andsalience.

Of course, if we are studying at the level of theword or sentence, we have access to vast databases ofreal language, known as corpora, with which to testour hypotheses. Corpus linguistics has become a majorresource for contemporary linguistics, supplanting thelimited sets of idealized or sampled text used in the past.So scholars studying multimodal sources also need accessto this kind of resource—for example, to test Kress andde Leeuwen’s proposed information values.

Bateman, Delin, and Henschel (2004) critique Kressand de Leeuwen and address the issue of how a multimodaldocument corpus might be constructed. Bateman’s (2008)major review of the field reveals some of the difficultiesto be faced. In particular, he notes that

A substantial set of problems is raised by the fact that theobject of study is not linear, either temporally or in terms ofthe principles for its consumption; moreover, its multichannelnature makes it difficult to reconcile and peg together themethods of recording, transcription, analysis and annotationthat have been developed separately for each mode. Thismakes empirical study and validation of theory particularlyproblematic: we have not had the “orderly arrangement ofthe objects upon which we must turn our mental vision” (cf.Descartes, Rule V of Rules for the Direction of the Mind,1701) and, partly as a consequence, analysis has remainedoverwhelmingly impressionistic. (Bateman 2008, 272)

There remains the question as to whether the annota-tion of material going into such corpora could ever beautomated, or whether it will always remain dependent on“impressionistic” analysis. Will computer pattern recog-nition ever be sophisticated enough to emulate the gestaltstructures seen by human readers, to use physiognomicrecognition, or to spot generic resonances—those visual,holistic features of a visual display that go beyond what isdefined in the markup languages (such as XML) that liebehind many modern documents? And will anyone ever

244 R. WALLER

think it worthwhile spending the time and money reachingthat goal?

A study by Thomas (2009; see also Thomas, Delin, &Waller 2010) illustrates the challenge of handling layout ina multimodal document corpus. A multimodal documentcorpus must allow users to search via the usual verbalstrings and tags, but also to search by layout propertiesand to view the actual document in facsimile. This meansa considerable amount of expert annotation at the inputstage, which discourages the development of large enoughcorpora to make computer analysis worthwhile.

LAYOUT IN THE DIGITAL AGE

Digital Genres

The genre approach asserts, in effect, that if there is aconsistent human need to communicate in a particularway, then a corresponding genre will probably emerge. Orif, as with letterpress printing, that need is not met, it willreemerge when technical developments allow it.

Although the digital age has made massively more doc-uments available to each user, it is still struggling to evolvea common basis for a digital literacy—an agreed set oftools and techniques that enable us to study, understand,and retrace our steps through information, as well as tofind it and connect it (both of which the digital age doessupremely well). It is an evolutionary process; for exam-ple, when publishers turned to interactive CD-ROMs inthe 1990s, they were rejected by the marketplace. Nowthose same publishers are tentatively moving back to sim-ilar formats for the Apple iPad, which provides a smootherdelivery path and a better user experience.

Paper documents often evolved through a process of ini-tial exploration and natural selection, followed by conser-vatism as production patterns settled and formats becamefamiliar. Document genres—categories such as “book,”“newspaper,” or “leaflet”—typically evolved out of thefunctional requirements of their producers and users. Evenwhen the original production constraints change, genressurvive because their users have conservative expecta-tions. So readers of novels expect serifed typefaces andstraight right-hand edges; readers of newspapers expectnarrow columns. Readers are conservative and want tominimize their effort to understand—genres represent asecure way in which writers can meet readers’ expecta-tions, and in which readers can understand the rules forreading any genre-conforming document. We know fromits format what the status of a paper document is, and whatto do with it: whether to keep it or discard it; whether toread it carefully or skim; even whether to treat it as enter-tainment, or as of serious import.

Something similar has to happen with digital docu-ments, and it will happen through creative explorations of

compelling interfaces and devices, rather than through thedeliberations of academic researchers. It has been notice-able that the popularity of tablet computers has spurred onmagazine publishers to experiment with innovative pagebased formats.

As a New York Times piece put it:

You’ve got to hand it to the magazine publishers. They con-tinue to throw spaghetti against the iPad and other e-readerstrying to see what will stick and what falls to the floor. (Bilton2010)

To date, most present their readers with defined, boundedcontent that, even when using interactive or video content,retains the page model and eschews the infinite exten-sion into hypertext and social interaction that characterizesmost newspaper websites.

Documents as Memory Tools

Digital genres are developing, and will continue to de-velop, particularly where they afford some functionalitythat was previously impossible. In particular, the digitalworld has not yet finished evolving usable formats formanaging the conversational dimension of websites. Thesame format is used for a handful of comments on a blogas for thousands of comments on a national newspaper’swebsite: a continuous scrolling page (sometimes slicedinto multiple pages), unsorted and unedited. Just as welearn to filter out irrelevant noise when have a conver-sation in a crowded room, we have to develop effectivefilters in order to find, read, and store online informationdiscerningly. As with other areas of human activity, wejust need the right tools.

Online channels give us incredibly effective tools whenwe are hunting for information. The question remains,though, whether we yet have the right tools for dissect-ing, cooking, and eating the information we have hunteddown. Conceptual thinking is about manipulating ideas.For example, with paper documents we might:

• Focus on an idea, integrating representations ofit from different sources—for example, a set ofdifferent documents, viewed together, open at rel-evant pages.

• Compare more than one document, point forpoint. People often do this by annotating docu-ments or transcribing concepts into tables or dia-grams.

• Park an idea, so that it is in view but not in play,to remember the fact of its existence. People typ-ically write notes for this, or leave books open ona desk.

• Connect a number of concepts, from inside andoutside of the document to hand. People sort


documents in piles, use color-coded bookmarks,or make lists and sketch diagrams.

• Prioritize among a set of possible directions forour thinking. People may transcribe ideas intonumbered lists, or sort documents into piles.

• Annotate a text element, to capture a thought be-fore it escapes. People underline or highlight doc-uments.

Some of these commonplace behaviors are still hardto achieve with the current generation of digital docu-ments. This is why Harper and Sellen (2002),10 in theirseminal study of document use in organizations, notedthat the default pose for many readers they observed (notjust writers) was with a pen in their hand, ready to an-notate the text or jot down new thoughts. In fact, theirremarkable conclusion was that in the digital age paper isprimarily an interactive medium, not a storage medium.The life cycle of a document, as they observed it, movesbetween digital and paper versions: marshaling and ex-tracting information (digital for search, paper for inte-grating multiple sources), writing (paper for planning,digital for drafting), editing/proofreading (paper), final-izing (digital), distribution and workflow (mostly digital),reading/consuming (paper better for longer documents),archiving/filing (digital). Paper is used by many as a tem-porary interactive medium: Documents are printed out forreading, annotation, comparing, and sharing—then recy-cled while remaining accessible in digital form. As theyput it:

We argue that we are not headed toward offices that use lesspaper but rather toward offices that keep less paper. Thisis because we will continue to need paper for some of thecritical work activities we do, but in these roles it will be verymuch a temporary medium. (209)

Of course, the judicious use of memory, and therefore ofmemory tools, involves forgetting as well as remember-ing. In a recent critique of the concept of “lifelogging”(the ultimate digital capture of every memory), Sellen andWhittaker (2010) suggest using psychological principlesfrom studies of human memory as design principles fordigital memories. Keeping the memory-tool metaphor inmind, this means designing digital formats that “strate-gically target the weaknesses of human memory.” Theyoutline what this means: for example, selective capture ofinformation, aiding metacognition and metamemory, anddesigning effective retrieval cues. This more or less de-scribes a book: curated, coherent, and designed to supportstrategic reading. But in the digital age this can be donein a way that is personalized, up-to-date, and configurableby the user.

Paper is not only essential for people wanting to spreadideas on the table and annotate them. It also has speed andcontinuity advantages over many current digital formats,

such as the Guardian spread in Figure 1. Its readers mustmove around not only with their eyes (as did Saenger’smedieval readers) but via trackpads and keyboards, andperhaps even search boxes, with resulting time delays andadditional cognitive load. Neilsen (1993) reported thattime delays during the use of interfaces causes readersto be distracted by a loss of fluency and loss of directcontrol.

In addition to the time delays, the physical position ofcontent in linear digital documents is usually not constant.This fluidity is also potentially disruptive, as there is ex-perimental evidence to back up a common observationthat readers use their memory of the physical location ofideas on a page when searching for previously read (see,e.g., Dillon 1991; Rothkopf 1971).11

Four Aspects of Digital Pages

Hypertext prophets used to speak as if the advent of digi-tal text were a paradigm shift, incommensurate with pastways of thinking and acting through text. It is perhapsmore common now to speak about the convergence oftechnologies and channels. In that spirit I identify fourpage archetypes that reflect generic resonances from thepast, the continuing need for traditional functions of thedocument, and the technical capabilities and connectnessof the current world (Table 1).

Fixed pages are the most diagrammatic. Because theyare locked in place, the reader can assume that relation-ships between elements (text blocks, pictures, headlines,etc.) are intentional and potentially meaningful. A pagebreak signifies the end of a unit of text, in the same wayas a sentence or a paragraph. The designer and writer, fortheir part, can craft graphic relationships, knowing thatthey will survive the various technical transmission pro-cesses and reach the reader.

Flowed pages are represented by traditional novels, orby e-reader books. The author’s words are flowed in andfill the pages one by one, with page endings that are asarbitrary as the line endings are. But those page endingsare fixed for the life of the document (or, in the caseof e-documents, until the text is reflowed after a changeof font). Readers can therefore move back and forth be-tween pages and use the constant geography of the book tonavigate.

Fugitive pages are formatted temporarily and perhapsalso populated with content temporarily. Pages are createdafresh for each reading, and may change when revisited.A common example is an online newspaper, which offersa reasonably coherent appearance and user experience butis constantly updated. If you return to a story later in theday, you may find that it has been relegated to a lowerposition in the hierarchy or even disappeared from view.Even if it is still there, the content may have been edited.

TAB

LE

1T

radi

tiona

land

emer

gent

page

type

sco

exis

ting

indi

gita

ldoc

umen

ts

Fixe

dFl

owed

Fugi

tive

Frag

men

ted

Defi

niti

onTe

xtan

dim

age

lock

edin

fixed

posi

tions

.L

inea

rte

xtflo

wed

into

page

tem

plat

es.

Page

sar

ecr

eate

ddy

nam

ical

lyat

poin

tof

use.

Page

sas

sem

bled

from

mul

tiple

sour

ces.

Typi

calp

urpo

seE

xpla

natio

n.N

arra

tive.

Com

mun

ity.

Enq

uiry

.

Exa

mpl

eM

agaz

ine.

Ref

eren

cebo

okde

sign

edin

page

sor

doub

le-p

age

spre

ads.

Tra

ditio

nalp

rint

edbo

ok.

E-r

eade

rbo

ok(e

.g.,

Kin

dle)

.O

nlin

ene

wsp

aper

.Se

arch

resu

lts.A

ggre

gato

rso

ftw

are

(e.g

.,Fl

ipbo

ard)

.

Impa

cton

cont

ent

and

layo

utC

onte

ntan

dla

yout

fixed

,al

thou

ghm

aybe

scal

edup

ordo

wn.

Con

tent

fixed

.Onc

eflo

wed

into

tem

plat

e,la

yout

isfix

edun

tilne

wed

ition

prod

uced

(or,

inth

eca

seof

e-re

ader

s,ne

wde

sign

para

met

ers

appl

ied

byus

er).

Page

scr

eate

daf

resh

from

data

+ru

les,

solo

oks

diff

eren

ton

each

chan

nelo

rde

vice

.

Page

scr

eate

dfr

omse

arch

syst

em,s

oas

sem

bled

from

mul

tiple

sour

ces.

Edi

tori

alco

ntro

lD

esig

ner

cont

rols

and

craf

ts,w

ithw

ords

edite

dto

fit.C

ompl

ete

cont

rolw

ithin

rest

rict

ive

canv

asco

nstr

aint

s(p

age

boun

dari

es).

Wri

ter

cont

rols

and

craf

ts,w

ithad

viso

ryan

dpr

oof-

read

ing

supp

ortf

rom

edito

r.M

inor

role

for

desi

gner

.U

ncon

stra

ined

bypa

gebo

unda

ries

.

Wri

ter

may

craf

tlin

ear

text

with

inco

nten

tman

agem

ent

syst

em.D

esig

ner

crea

tes

tem

plat

esfo

rte

xtto

flow

into

.L

ittle

cont

rolo

rcr

aftin

g.C

onte

ntan

dla

yout

man

aged

thro

ugh

prog

ram

med

rule

s.

Rea

der

sets

and

test

sru

les.

Litt

lecr

aftin

gor

edito

rial

cont

rol,

exce

ptth

roug

hst

rate

gies

such

asse

arch

-eng

ine

optim

izat

ion.

Rea

der

cont

rol

Loo

kan

dre

ad.

Usi

ngge

stal

tper

cept

ions

ofst

ruct

ure.

Mov

ing

from

foca

lto

peri

pher

alaw

aren

ess

ofst

atic

info

rmat

ion

tovi

ew,a

ndpl

ace

mem

ory

ofin

form

atio

nal

read

yvi

ewed

.

Rea

d.B

uild

ing

men

tal

repr

esen

tatio

nof

stru

ctur

e.R

eadi

ngst

rate

gica

llyto

chec

kan

dre

info

rce

thos

est

ruct

ures

.Mov

ing

back

and

fort

hbe

twee

nst

atic

page

s.

Clic

kan

dre

ad.

Sim

ilar

toflo

wed

text

,but

with

less

relia

bilit

yw

hen

revi

sitin

gpr

evio

usly

foun

din

form

atio

n.

Clic

k,re

ad,c

lick,

read

.B

uild

ing

men

tal

repr

esen

tatio

nfr

omm

ultip

leso

urce

s.So

phis

ticat

edle

vels

ofin

fere

nce

need

edto

inte

grat

efr

agm

ents

.

246


Fragmented pages are compilations of page elementsfrom a variety of sources that may not have any rela-tionship predictable by their authors. An example is theresults of a search, or an aggregation application such asFlipboard (which assembles content from a range of theuser’s favorite sources, such as blogs or social networkingsites, into a magazine-like format).

These page types may exist in pure form or coexist incombination. For example, an online newspaper may havefixed layouts into which fugitive content is flowed, and acolumn of fragmented advertisements drawn in throughpersonalization rules.

Illustrated textbooks and catalogues are a common hy-brid of fixed and flowed text: Pictures or marginal notesneed to be related to specific points in the text, so twocolumns may flow together through the book.

LAYOUT AND GRAPHIC LITERACY

I have discussed the relative difficulty in delivering fixed,laid-out pages via digital channels that are designed todeliver flowed, fugitive, and fragmented text. I now con-sider a different kind of barrier: the skills needed to pro-duce effective layout, and the low priority that it is givenwithin many information-providing organizations. I usethe term “graphic literacy,” although this is to extend aterm that is more often associated with the ability to inter-pret pictures and charts.12

In everyday usage the term “literacy” usually refersboth to the ability to read and to the ability to write prose.Most Western countries have very high rates of proseliteracy—around 99% is typically claimed—but muchlower rates of functional literacy:

[E]ven the most economically advanced societies have a lit-eracy skills deficit. Between one-quarter and three-quartersof adults fail to attain literacy Level 3, considered by ex-perts as a suitable minimum skill level for coping with thedemands of modern life and work. (OECD 2000, xiii)

As we have seen, adult literacy tests go beyond proseliteracy to include document literacy and quantitative lit-eracy. Tests of document literacy claim to measure theability to use complex documents (which include manydifferent text features in addition to continuous prose) inorder to do tasks that involve departing from the linearstructure to search, compare, make inferences, and solveproblems.

So we might construe graphic literacy (or more specif-ically, typographic literacy) as the key difference betweenprose literacy and document literacy. In other words: doc-ument literacy = prose literacy + graphic literacy.

The documents used in tests of document literacy in-clude forms, timetables, instructions, and other everydayfunctional documents. Some of these are highly conven-

tionalized, and in those cases, literacy must thereforeinvolve familiarity with conventions typical found inparticular document genres. Hamilton and Barton (2000)criticized the IALS test, as used in the United Kingdom,on exactly this point (among others):

Looking more closely, there are US ways of using languageand US conventions of design and layout. The bus timetable,for instance, follows the twelve hour clock with a.m. andp.m. The morning is written in normal font and the after-noon in bold. This is fine, it is comprehensible and it mayseem innocuous. Nevertheless, these are US conventions;in Britain bus timetables normally use a twenty-four-hourclock; morning and afternoon buses are not given a differentfont. Font differences are usually used to distinguish throughservices from ones where a change of bus is required. Theseare small points, but they are indicative of how the seeminglyculture-free bus timetable may in fact be quite a different textin two countries and be clearly perceived by respondents asoriginating outside of their own culture. (383)

It is arguable that to be unaware, as a literacy test designer,of the localized nature of document genres is itself a formof graphic or document illiteracy, tantamount to not real-izing that in other countries they speak foreign languages.

The test question for the document in Figure 7 is: “Sup-pose the annual budget statement will be 105 pages andyou need to distribute 300 copies. Would Quick Copy dothis job? Explain your answer.” The document is poorlydesigned on several levels: It fails to use well known genrerules, it fails to use layout and design features to direct at-tention and afford effective use, and its content relates to ahighly local and specific system. If a user fails this literacytest, whose literacy is lacking: that of the user or of thedocument creator?

Hamilton and Barton are key figures in the “new litera-cies” movement, where specific literacies are identifiedamong different discourse communities (usually called“situated literacies,” but for clarity in this context I callthem “conversational literacies”). They define this ap-proach in the same article:

Our approach is based upon a belief that literacy only hasmeaning within its particular context of social practice anddoes not transfer unproblematically across contexts; thereare different literacy practices in different domains of sociallife, such as education, religion, workplaces, public services,families, community activities; they change over time andthese different literacies are supported and shaped by thedifferent institutions and social relationships. (Hamilton &Barton, 2000: 379)

So ideally, then: document literacy = prose literacy +graphic literacy + conversational literacy.

Conversational literacy describes our understanding, ei-ther as creators or users of communication channels, ofhow a particular communication is shaped by its conver-sational context. It recognizes that each participant brings

248 R. WALLER

FIG. 7. An example of a level 3 IALS test item, from the Literacy Task Assessment Guide, National Literacy Secretariat, HumanResources and Skills Development Canada (Evetts & Gauthier 2005, 114).

his or her own motives and experience to a conversation,and it understands how any document is likely to be inter-preted in a particular context and the range of inferencesthat it is reasonable for readers to make.

Producers and Consumers: An Imbalance of Accessand Skills

When discussing document or digital literacy, we some-times forget that traditionally we speak of literate peopleas being able to both read and write. Applying the sameprinciple to document literacy, this means that a failureof communication may be blamed on the literacy skills ofboth producer and user.

The most sophisticated written documents are producedby individuals who have highly developed skills of writ-ing, editing, and design, working in industrialized systemsof production and distribution. Their skills are held in theform of procedural knowledge developed among com-munities of practice, and learned through apprenticeship,rather than declarative knowledge taught through formalgrammars.

Page-based (normally, paper) channels, in their mostevolved forms, involve an imbalance in the access to com-munication channels of elite producers (in the form ofauthors and publishers) and consumers. Digital channelsare now open to all, but the skills required to communicateeffectively are not universal.


FIG. 8. An official notice about a traffic violation (from Waller2012). Original size A4.

The problem of competence is a key problem whenwe consider layout as an aspect of text that carries mean-ing. When linguists study spoken language by recordingspeakers, they accord them respect by attributing vari-ations from “standard” forms to such things as dialect,mood, or context. But when we study written language,and in particular typographic layouts, it is hard to avoiddistinguishing between trained and untrained writers anddesigners. Although they are typically called “expert” and“lay” designers by researchers (see, e.g., Walker 2001),if they were to be considered equally competent, therewould be no such profession as graphic designer.

How, then, might we extend the concept of relative doc-ument literacy to the document producer? Fully document-literate document creators (whether persons or organiza-tions) must work at three corresponding levels, measurablethrough user testing:

Prose level: They must be able to write fluent, andreadable prose—the traditional criterion for an educated,literate person.

FIG. 9. A redesigned version of the Penalty Charge Notice (aspeculative draft by the author, not implemented).

Graphic level: They (if necessary collaborating with adesigner, or using templates) must be able to use layout andtypography to create a usable environment for searching,skimming, and seeing content structures diagrammed, aswell as the close reading of prose. We could go furtherand say that they must be able to use alternatives to prose,such as pictures, diagrams and charts, where these wouldbe more effective.

Conversational level: They must create an encounterof user and document in which a range of appropriate be-haviors, reader roles, and critical stances is made obvious,and in which key prior knowledge or postreading actionsare made plain.

An example might help to explore the distinction be-tween the graphic and the conversational levels, shown inFigure 8. The official document in Figure 8 fails on allthree counts:

250 R. WALLER

• Prose: It is written in bureaucratic language un-suitable for a general audience.

• Graphic: Its structure is poorly articulated graph-ically, with numerous ways to structure and high-light information used simultaneously.

• Conversational: It does not make its functionclear, or the process of which it is a part.

In some ways, this documents reflects the IALS test inFigure 7. It shows a similar lack of design competence,and is similarly adrift from the range of everyday genresthat readers are used to. Successful readers of either doc-ument have to imagine a possible world in which it makessense, and within which they can make inferences aboutintention—in much the same way as when reading an un-grammatical sentence, we try different meanings until oneof them appears to makes sense in context.

The redesigned version of the Penalty Charge Notice(Figure 9) has a clear visual pathway that corresponds toan explanation pathway: This is what happened, here isthe proof, this is what to do next, and this is how to do it.

Ideally, document designers can take an existing genreas a model—something shared with their readers in agiven discourse community. Official documents like thisone, however, are encountered too rarely in the lives ofindividual to establish strong genre conventions. One ofthe problems with the original version is that it showsevidence of excessive repair: over-signaling is often theresult of attempts to overcompensate for poor readerresponses.

In the absence of a strong genre, the new design fallsback on the core techniques of clear information design:The content is organized as a narrative, told in the left-hand column, with very clear framing (the horizontalrules). What happened (and the evidence); what is thepenalty (and how much); then finally a choice of ap-peal or pay. With a pattern language perspective, it fol-lows a strong procedural “action and result” pattern foundmore frequently in user guides. This kind of layout wouldnot be out of place in a quick start guide to tell usershow to install an ink cartridge in a printer or how toprogram a digital watch. The payment area at the footof the page borrows a payment slip pattern from utilitybills.

In this section I have tried to place layout at the heart ofdocument literacy and communication competence. Tra-ditionally, page layout is the province of specialist graphicdesigners, who are normally employed only on a limitedrange of documents. But if what they do makes an impor-tant contribution (and I believe it does), then it deservesto be seen as a core communication competence that ev-ery communicator shares, that every communication toolenables, and that every student of textual communicationrecognizes.

CONCLUSION

Information, knowledge, message, and document: Eachword brings its own personal, social, and technical per-spective.

In our discussions of knowledge management in thedigital age (the context in which this paper was orginallypresented), we should not forget that documents are morethan linear text. They are multimodal juxtapositions ofelements whose spatial relationship may be every bit asintentional, essential, and effective as the order of words insentences. Making documents, transmitting them, archiv-ing them, repurposing them, integrating them into the so-cial context of the connected digital world: These are chal-lenges in which the subtleties of crafted displays are easilylost while we focus on the newer technical challenges in-volved in managing large numbers of documents, and intracking complex conversations.

NOTES

1. Among others, by thinkers such as Ong (1982) on the move fromorality to literacy, Eisenstein (1979) on the move from manuscript toprint, and McLuhan (1962) on the move from print to television. Baron(2008) has reported how the online world is changing our use andexpectations of language.

2. Fortunati (2010) argues that the involvement of so many of oursenses in the reading of paper documents provides a memory supportnot present in electronic reading.

3. A commutation test, in semiotics, tests the relative strengthof a potential signifier by changing it in some way—for example, byremoving or substituting it in order to assess its contribution to meaning.See Chandler (2007) for a fuller account.

4. For example, the Inkling format (www.inkling.com) and theApple iBooks textbook (www.apple.com).

5. In her foreword to Steinberg (1974).6. Collectively we can describe these as “access structures”

(Waller 1979). Research on their educational effectiveness is reviewedby Britton and Black (1985).

7. De Beaugrande (1984) presents a typology of linearization inlanguage, which includes what he terms “core-and adjunct” (which in-cludes contrast, and hierarchical relationships), pause, heaviness, list-ing, disambiguation, looks-back, and looks-ahead.

8. I developed this analogy further in Waller (1982).9. See White (1982) for insight into the magazine designer’s craft.

10. Although at the time of writing, this study is already 10 yearsold, in 2002 electronic documents, e-mail, intranets, and the Web werealready well established in all large organizations.

11. I wrote a short critique of hypertext in its early days, entitled“What electronic books will have to be better than,” which highlightedthe role of physical constancy in enabling intensive study, using anactive reading strategy (Waller 1986).

12. The International Visual Literacy Association (IVLA) acknowl-edges the existence of a multitude of definitions of visual or graphicliteracy, remarking on their website that each scholar has produced hisor her own (www.ivla.org/org what vis lit.htm, accessed February 12,


2012). Brill, Kim, and Branch (2007) attempted to reach agreementamong experts, with limited success.

REFERENCES

Adobe Systems, Inc. 2010. eBooks: Common questions aboutcreating EPUB files with Adobe InDesign. http://www.adobe.com/products/indesign/epub/howto/pdfs/eBooks common questions.pdf(accessed August 1, 2010).

Anderson, T., and B. Armbruster. 1985. Studying strategies and theirimplications for textbook design. In Designing usable texts, ed. T.M. Duffy and R. Waller, 159–77. Orlando, FL: Academic Press.

Baker, L., and A. Brown. 1984. Metacognitive skills and reading. InHandbook of reading research, ed. P. D. Pearson, 353–94. New York:Longman.

Bateman, J. 2008. Multimodality and genre: A foundation for the sys-tematic analysis of multimodal documents. London, UK: PalgraveMacmillan.

Bateman, J., J. Delin, and R. Henschel. 2004. Multimodality and em-piricism: Preparing for a corpus-based approach to the study ofmultimodal meaning-making. In Perspectives on multimodality, ed.E. Ventola, C. Cassily, and M. Kaltenbacher, 65–87. Amsterdam:John Benjamins.

Berger, S. 2005. Allotment gardening. Totnes, UK: Green Books.(Kindle edition 2010).

Bernhardt, S. 1985. Text structure and graphic design. In Systemicperspectives on discourse, ed. J. Benson, and W. Greaves, 18–38.Norwood, NJ: Ablex.

Bilton, N. 2010, June 29. Digital magazines don’t encouragesocializing. New York Times. June 29. Available at http://bits.blogs.nytimes.com/2010/06/29/digital-magazines-dont-encourage-socializing/

Black, C., ed. 2006. Transformation of knowledge: Early manuscriptsfrom the collection of Lawrence J Schoenberg. London, UK: PaulHolberton Publishing.

Bouayad-Agha, N., D. Scott, and R. Power. 2001. The influence oflayout on the interpretation of referring expressions. In Multidis-ciplinary approaches to discourse, ed. L. Degand, Y. Bestgen, W.Spooren, and L. van Waes, 133–41. Munster: Nodus Publikationen.

Brill, J. M., D. Kim, and R. M. Branch. 2007. Visual literacydefined—The results of a Delphi study: Can IVLA (operationally)define visual literacy? Journal of Visual Literacy. 27: 47–60.

Brown, A. 1980. Metacognitive development and reading. In Theoret-ical issues in reading comprehension, eds. R.J. Spiro, B. Bruce, andW. Brewer, pp. 453–479. Hillsdale, NJ: Lawrence Erlbaum Asso-ciates.

Chandler, D. 2007. Semiotics: the basics, 2nd ed. London, UK: Rout-ledge.

Delin, J., J. Bateman, and P. Allen. 2002. A model of genre in documentlayout. Information Design Journal 11: 54–66.

Edwards, P. A. 2010. Reconceptualizing literacy. Reading Today 27:22.

Eisenstein, E. E. 1979. The printing press as an agent of change.Cambridge, UK: Cambridge University Press.

Evetts, J., and M. Gauthier. 2005. Literacy task assessment guide.Ottawa, Canada: National Literacy Secretariat.

Farkas, D. K., J. Larson, and S. J. Naranjo. 2011. LabelPatterns.org: Acomprehensive pattern library for consumer-decision labels. IEEE

International Professional Communication Conference. Cincinatti,Ohio.

Farthing, F. Hadfield. 1929. Everyday in my garden. London, UK:Knopf.

Foltz, P. W. 1996. Comprehension, coherence and strategies in hyper-text and linear text. In Hypertext and cognition, ed. J.-F. Rouet, J.J.Levonen, A. Dillon, and R. J. Spiro, pp. 109–136. London, UK:Routledge.

Forceville, C. 2007. Review of Anthony Baldry and Paul J.Thibault, Multimodal transcription and text analysis: A multime-dia toolkit and coursebook. Journal of Pragmatics. 39: 1235–38.

Fortunati, L. 2010. Some thoughts about electronic texting. Paper pre-sented at COST/European Science Foundation Strategic Workshop“Electronic Textuality,” June, Istanbul.

Gamma, E., R. Helm, R. Johnson, and J. M. Vlissides. 1994. Designpatterns: Elements of reusable object-oriented software. Reading,MA: Addison-Wesley.

Halliday, M., and R. Hasan. 1976. Cohesion in English. London, UK:Longman.

Hamilton, M., and D. Barton. 2000. The international adult literacy sur-vey: What does it really measure? International Review of Education46: 377–89.

Herrlinger, R. 1970. History of medical illustration, from antiquity toA.D. 1600. London, UK: Pitman Medical.

Hessayon, D. G. 1967. The rose expert. London, UK: Trans-WorldPublishers.

Hjelmslev, L. 1959. Essais linguistiques. Copenhagen: Nordisk Sprog-og Kulturforlag.

Kinross, R. 1992. Modern typography: An essay in critical history.London, UK: Hyphen Press.

Kress, G., and T. van Leeuwen. 1996. Reading images: The grammarof visual design. London, UK: Arnold.

Landow, G. P., and P. Delany. 1991. Hypertext, hypermedia and lit-erary studies: the state of the art. In Hypermedia and literary stud-ies, ed. P. Delany & G. P. Landow, 3–50. Cambridge, MA: MITPress.

McKenzie, D. 1986. Bibliography and the sociology of texts (ThePanizzi Lectures, 1985). London, UK: The British Library.

McLuhan, M. 1962. The Gutenberg galaxy. London, UK: Routledge& Kegan Paul.

Nielsen, J. 1993. Usability engineering. San Francisco, CA: MorganKaufmann.

Organization for Economic Cooperation and Development. 1997. Lit-eracy skills for the knowledge society. Paris, France: OECD.

Organisation for Economic Co-operation and Development (OECD).2000. Literacy in the information age: Final report of the Interna-tional Adult Literacy Survey. Paris, France: OECD.

Paris, S. G., and M. Myers. 1981. Comprehension monitoring, memory,and study strategies of good and poor readers. Journal of ReadingBehavior 13: 5–22.

Pennac, D. 2006. The rights of the reader. London, UK: Walker Books.Pettersson, R. 2007. Visual literacy in message design. Journal of Visual

Literacy 27: 61–90.Polanyi, M. 1969. Knowing and being. Chicago, IL: University of

Chicago Press.Power, R., and D. Scott, eds. 1999. Using layout for the generation

understanding or retrieval of documents: Papers from the AAAI FallSymposium. Palo Alto, CA: AAAI Press.

252 R. WALLER

Pugh, A. 1975. The development of silent reading. In The road toeffective reading, ed. W. Latham, pp. 110–119. London, UK: WardLock.

Rothkopf, E. 1971. Incidental memory for the location of informationin text. Journal of Verbal Learning & Verbal Behavior 10: 608–13.

Saenger, P. 1982. Silent reading: its impact on late medieval script andsociety. Viator 13: 367–414.

Schriver, K. 1996. Dynamics in document design: Creating text forreaders. New York, NY: Wiley.

Sellen, A., and S. Whittaker. 2010. Beyond total capture: A constructivecritique of lifelogging. Communications of the ACM. 53: 70–77.

Steinberg, S. 1974. Five hundred years of printing, 3rd ed. Har-mondsworth, UK: Penguin.

Sutnar, L. 1961. Visual design in action: Principles, purposes. NewYork, NY: Hastings House.

Sutnar, L., and K. Lonberg-Holm. 1944. Catalog Design. New York,NY: Sweet’s Catalog Service.

Thayer, A., C. P. Lee, L. H. Hwang, H. Sales, P. Sen, and N. Dalal. 2011.The imposition and superimposition of digital reading technology:the academic potential of e-readers. In CHI 201: Reading & writing,2917–26. Vancouver, BC: ACM.

Thomas, M. 2009. Localizing pack messages: A framework for corpus-based cross-cultural multimodal analysis. PhD dissertation, Univer-sity of Leeds, Leeds, UK.

Thomas, M., J. Delin, and R. Waller. 2010. A framework for corpus-based analysis of the graphic signalling of discourse structure. Pa-

per presented at Multidisciplinary Approaches to Discourse, March,Moissac, France.

Tidwell, J. 2005. Designing interfaces: Patterns for effective interactiondesign. Sebastopil, CA: O’Reilly.

Walker, S. 2001. Typography and language in everyday life. London,UK: Longman.

Waller, R. 1980. Graphic aspects of complex texts: Typography asmacro-punctuation. In Processing of visible language, ed. P. Kolers,M. Wrolstad, and H. Bouma, pp. 241–253. New York, NY: PlenumPress.

Waller, R. 1986. What electronic books will have to be better than.Information Design Journal 5: 72–75.

Waller, R. 1987. Typography and reading strategy. In Executive controlprocesses in reading, ed. B. Britton, and S. Glynn, 81–106. Hillsdale,NJ: Lawrence Erlbaum Associates.

Waller, R. 2012. Simple action 1: Penalty charge notice. London, UK:The Simplification Centre.

Waller, R., and J. Delin. 2010. Towards a pattern lan-guage approach to document description. Paper presented atMultidisciplinary Approaches to Discourse, March, Moissac,France.

Wertheimer, M. 1938. Laws of organization in perceptual forms. InA source book of Gestalt psychology, ed. W. Ellis, 71–88. London,UK: Routledge & Kegan Paul.

White, J. V. 1982. Editing by design, 2nd ed. New York, NY:Bowker.

Copyright of Information Society is the property of Routledge and its content may not be copied or emailed to

multiple sites or posted to a listserv without the copyright holder's express written permission. However, users

may print, download, or email articles for individual use.

Date post:	21-Dec-2014
Category:	Design
Upload:	asliza-hamzah
View:	84 times
Download:	1 times