This article first appeared in
The Reader: Journal of the Arkansas Reading Association, Summer, 2001, a peer-reviewed journal, reprinted here with permission. Since its publication, I've studied Jakob Neilsen's work; please read
Critique of "How Users Read on the Web."
Reading on the Web
While others were playing Mario Brothers, Myst and Sim City, I was playing webpage. Since I considered it play, I didn't learn fast and I played by my own rules. Along the way, I've read widely and experimented with what it means to write for the web. Like others, I'm just learning to intelligently discuss the nuances of web page design. Equally fascinating is the
emerging data on how people read on the web. Once we have enough data on how people actually use the web, we will be able to plan teaching strategies that will help children use the web efficiently and pleasurably.
Basic Definitions
First, we need to make sure our
vocabulary is clear.
Hypertext is the linkage of one piece of information to another piece of information. A cross-referenced Bible is a print example of hypertext. Hypertext on a webpage, though, is a more fluid experience. Doubleclick a blue-highlighted word and it takes you to a new page, a new visual experience.
While the entire internet is linked, and thus hypertext, there are different types of linked pages. First, is a piece of
hypertext narrative (essay, fiction, poem) which is a group of pages that should be read together to create a coherent thought or story; these pages generally don't link to any outside information except at the beginning or the end.
A
website is a group of pages, usually authored by the same person or group, that can include hypertext narrative, but can also cover a number of topics. For example, on my
website,
, I have a variety of topics: notes from a teacher in-service on teaching writing, archive of reviews of children's books first published in the Arkansas Democrat-Gazette, information about my books, and a hypertext narrative essay about writing The Wayfinder.
Finally, there is the internet, millions of pages worldwide, which are randomly linked.
Strategies for Web Design
Currently, there are three strategies for designing websites or webpages: visual, informational, narrative. Visual web design emphasizes the graphic capabilities of hypertext, including Flash programing or other animated graphics.
Informational web design emphasizes the non-fiction, information-rich strategy to web design; visually, these pages are minimalistic. The main design criteria for informational web design is usability. Can the user find the information he needs and is it in a form he can use? Web designers who hold this point-of-view consider themselves Information Architects.
Narrative web design emphasizes the story or the process of story that occurs as a reader journeys through a website. Visual aspects aren't as important as the mental connections made from page to page. Facts can be presented, as in any good creative non-fiction, but it is the story that controls the presentation of facts.
These three approaches can be visualized as the corners of a triangle; within the triangle fall all webpages. Rarely is any webpage or website a pure expression of one strategy; rather the continuum among the approaches can be expressed in a multitude of ways. When you teach children to use the web, they should have experiences with all three types of webpages and start to understand the strengths and weaknesses of each.
Teaching children to read on the web does include visual aspects, but because text is secondary in this design strategy, I won't deal with it here. Reading information webpages is different from reading narrative webpages; each require a different set of skills and different strategy for understanding.
READING INFORMATIONAL PAGES
Jakob Nielsen, one of the leading proponents of usability as the criteria for good web design, recently wrote about skills needed to use the web effectively. In his article, The 3Cs of Critical Web Use: Collect, Compare, Choose", he suggests three categories of skills to teach.
Collecting Information
Before you can start to collect information, you need to understand how human physiology affects our reading on the web. Edward Tufte has said that when displaying visual information, the human eye is designed specifically to decipher visual complexity; however, human's visual memory is poor.
For example, Yahoo search engine displays numerous choices on its front page. The visual hierarchy and the presentation of multiple columns compresses the information into a single screen which is visually easy to understand and use. The alternative, of listing keywords in a single column, would require that the reader scroll through several screens. Because visual memory is not as reliable, that design would make Yahoo difficult to understand and use.
Tufte uses layers of information, often clarified by the use of color, to create pages of complex, yet usable information. This suggests that children should be taught to decipher increasingly complex sets of graphic data, including lists, charts, graphs, or flowcharts.
A second problem is how current hardware affects reading on the web. With the screen resolution of today's monitors, people read 25% slower on the internet than hard copy. (Nielsen, "How Users Read on the Web") Rarely is the entire text read; instead, 79% of readers scan for important information. Writers are advised to condense material, to make bulleted lists and to highlight important words. If most adult readers scan for information, then the ability to scan quickly and correctly is another skill children need to learn.
Nielsen's research also reports that 71% of users search for multiple pieces of information. They are driven by a specific goal, but are not looking for one particular answer. 96% of users are looking for specific information.
Nielsen's book discusses in detail how to construct a website to allow readers to find the specific information they need. About 50% of users are search-dominant while 20% are link-dominant and the rest are mixed-dominant. (Neilsen 224) These are two strategies for finding information on a website. Search-dominated people immediately key in to the search engine for a certain page. Link-dominant people will use the navigation keywords to guess where the information might be located. Mixed-dominant people will start as link-oriented, but if that's not successful within a few clicks, they will turn to a search engine. While Nielsen uses this information to suggest, for example, that search engines be standardized in the top right corner of a webpage, it also has implications for teaching reading.
Readers, then, need to learn to use search engines and the conventions of these, such as the use of the plus or minus sign to force an engine to include or exclude a word. Another example is using quotation marks to enclose a phrase to force the engine to look for the entire phrase, not just separate words. Basic tutorials on search engines make search-oriented users more effective.
Link-dominant readers need to learn to anticipate what they will find if linking on certain keywords. Study typical keywords and learn what might the keyword typically represents. For example, FAQS refers to "Frequently Asked Questions." Mixed-dominant readers need to learn both strategies.
A final step in collection of information is the format for collecting: hardcopy or bookmarking a webpage. Readers need to learn when to read on-line and when to print out a page. If the text is small, no words are highlighted or bulleted, subtitles are missing–if the writing isn't designed for easy scanning–but it looks like information you need, print it. It will be much easier to read as hardcopy.
Comparing Information
Once several pages have been identified as potentially providing needed information, you must compare the pages. Factors to compare include these:
- Who has written this page? Teach the difference among .edu, .gov, and .com sites. Is the author authoritative on the topic?
- What sort of information is provided? Opinions, facts, statistics, etc.
- How reliable is the information?
Choosing Information
readers will have criteria for deciding what information to use. The criteria are often intuitive or unspoken. Students need to learn to articulate these criteria and how the criteria might vary for different needs.
For example, my sister and worked on a manuscript, How Healthy is Your Environment?. I wanted to include information on the effects of acid rain on the Black Forests of Germany, specifically, what percentage of the forest was damaged by acid rain. Internet sources reported a range of damage from 30-70%. Who was right? None of the webpages references scientific articles: I rejected all the webpages as unreliable for my purposes. I went to the library and found a book about acid rain copyrighted in 1970, about the time of the initial reports of Black Forest damage. The book referenced the original scientific article. My criteria was that the data must be backed up with scientific research.
READING NARRATIVE WEBPAGES
Hypertext narrative essays, poems, or fiction have organization scheme that we don't readily recognize from print. Print literature is in a fixed relationship; one page follows the next, which follows the next. You don't see page one followed by page 325. Hypertext presents information in relative order, which is partially determined by the reader. (The term used for one screen of information is lexia. A hypertext fiction might have 85 lexia with 225 links; it's a different way of describing length of a piece of literature!)
As a writer of fiction, I'm struggling to see how this can work in a story. Traditional stories build to climaxes. In hypertext fiction, the reader could choose to read the climax first, then any of the pages in any order. Fortunately, new strategies of narrative organization are emerging.
In her book, Radical Change: Books for Youth in a Digital Age, Eliza T. Dresang discusses the changing look of young adult literature, including new organizational structures. These "radical change" books are remarkable similar to the structures emerging in hypertext narrative, but on a different scale: instead of chapters, a single webpage or collection of a few webpages may represent one section of the structure.
For example, Paul Fleischman's book, Whirligig, is about a boy who tries to commit suicide by crashing his car head-on into another car. The boy survives, but kills the driver of the other car. As penance, the victim's mother says he must build and set up a whirligig in each of the four corners of the United States. The story isn't presented in strict time order. Instead, the four whirligigs are presented out of order and often the time is changed. The context may be ten or twenty years after the whirligig is set up. Yet, from this non-linear organizational structure, the reader is able to mentally construct an accurate time line and understand the impact of the whirligigs on the main character and on others who encounter them.
Hypertext narratives can present altered time-lines, can manipulate contexts and many other non-linear strategies. It would be helpful for children reading hypertext narratives to first read and discuss the organization of "radical change" books.
Organization Schemes for Hypertext
Mark Bernstein's article, "Patterns of Hypertext," describes ways to organize hypertext.
Bernstein identifies four different types of cycles, which "create recurrence." For example, three Billy Goats each take turns crossing a bridge, but with different strategies. Sometimes the repetition is exact, sometimes varied.
Counterpoint involves two voices which "alternate, interleaving themes or welding together theme and response." An example would be telling a baseball story by alternating the pitcher's and the batter's points of view. Similar is a mirrorworld, which "provide a parallel or intertextual narrative that adopts a different voice or contrasting perspective." Add in a baseball coaches narrative to the pitcher's narrative and you'll see different points of view.
Tangles provide a reader numerous links "without providing sufficient clues to guide the reader's choice." This can be one of the most disorienting of structures, but it can also be amusing; they encourage "browsing and discovery."
Sieves are like branching trees such as the Yahoo search engine home page (http://www.yahoo.com), which allows you to choose topics, which lead to a group of pages on that topic.
In a montage, "several distinct writing spaces appear simultaneously, reinforcing each other while retaining their separate identities." Think of the front page of a newspaper, where different stories are presented at once.
Neighborhoods establish "an association among nodes through proximity, shared ornament, or common navigational landmarks." Usually these are pages grouped together from one website, which should be considered together. Unique visual markers often indicate the presence of a neighborhood.
Split and Join pattern "knits two or more sequences together." One point of view takes off on a tangent, but eventually rejoins the main narrative. The first pig goes off to build his house of straw, but eventually the story comes back to all the pigs.
Missing Links are identified when the hypertext suggests "the presence of a link that does not, in fact, exist." Another navigational strategy is a feint. It "establishes the existence of a navigational opportunity that is not meant to be followed immediately; instead, the Feint informs the reader of possibilities that may be pursued in the future." You might consider this foreshadowing.
For an interesting discussion of how the cycles of baseball might illustrate these organizational strategies, see Bernstein's essay "Hypertext Narrative and Baseball." For another interesting connection, read the essay, then study the structure of Virginia Euwer Wolff's middle-grade novel, Bat 6 (Scholastic, 1998).
Patterns in Use: "Finding My Way Into Fantasy" essay
After reading and thinking about these organization schemes, I decided to try it with non-fiction. I took a speech about how I wrote The Wayfinder and "translated" it into hypertext using the StorySpace program from Eastgate.com. The resulting hypertext essay is very much a translation, with 32 lexia, 61 links. If you hear me give the speech, it is a very different experience from reading the hypertext version. The most common strategy used is cycles of various types and sizes, but there are also navigational feints.
Hypertext narratives and information webpages are the most common types of reading that children will encounter on the internet. These strategies for teaching reading on the web are merely a beginning point. As the internet evolves, we'll need to continue to analyze what is happening and how to best help children read effectively from the web.
Text ©2002/2003 Darcy Pattison. All rights reserved.