Scrolling is the single gesture–besides maybe walking or blinking–that I perform most in my everyday life. Two fingers, or maybe just a thumb, sliding across frictionless glass, panning the screen downward. Just looking at your fingers, it seems like you’re indulging in an anxious tick, or furtively shepherding the data out of your machine. As automatic as breathing, as pervasive as air, it has become hardwired into our metacarpal musculature. Research indicates that most users begin scrolling immediately—sometimes before a page has even loaded. Scrolling is so ingrained as to seem part of the natural order of our interfaces, but it is no more natural than any of its potential alternatives. The history of scrolling reveals several internets that could have been, each resting on a different solution to the simple problem: How to fit all of that data on one screen?
Genealogy of scrolling
There are two different frameworks through which one can understand the lineage of scrolling. The first treats scrolling primarily as a technology for reading, comparing it to other historical reading interfaces such as the wax tablet, the codex, or its eponymous paper scroll. The second frames scrolling as a visual phenomena, comparable to other developments which have occurred on the picture-plane in Western visual culture. As the erratic blend of image and text on any contemporary newsfeed demonstrates, neither approach is fully sufficient. Everything that can happen on a screen is within scrolling’s purview, and its mechanics are indifferent to the type of content displayed. Scrolling is a fundamentally visual phenomena in that its basic action is a simple geometric transformation, yet its function is identical to that of the page turn and other physical gestures that retrieve information; this becomes obvious when you consider that it has largely replaced them.
Scrolling as seeing
Many elements from the Western artistic canon have left their traces in computer interfaces. Both the screen and the painting are flat, rectangular surfaces which contain an enframed virtual space. Furthermore, as in painting, scrolling on a computer screen requires a certain suspension of disbelief: the flat plane you are gazing at must be understood as a virtual space that can be navigated. This is best illustrated by a classic debate surrounding scrolling: Are you, the viewer, moving your perspective through an imaginary space? Or is your perspective fixed while the document moves in front of you?
This question dates back to the early 1980s, when the Xerox Star chose inward-facing arrows for its scrollbar, while its competitor, the Apple Lisa, chose outward-facing arrows. In each case, the mechanism is exactly the same—the top arrow scrolls up, the bottom arrow scrolls down—but the premise is inverted. Xerox’s scrollbar proposes that its handles control the document, which you are pulling downward when you scroll up, and upward when you scroll down. Apple’s scrollbar implies that you are in control of your own perspective, which you can move along a rigid path. Although desktop computers tended to adopt the Apple scrollbar, smartphone interfaces moved toward the Xerox system. Touchscreens allow us to push and prod rectangular frames—even, sometimes, to fling them offscreen.
Apple’s 2011 OS X Lion, in turn, imported another of the smartphone’s scrolling characteristics back to the desktop environment: it made the scrollbar invisible. On smartphones, there is often no on-screen indication of scrolling other than its effect. Your thumb is doing the visual labor here. For users of an Apple desktop or device, the scrollbar is now ghostly. It might sometimes materialize, briefly, but always fades out of view as soon as the gesture closes. Its default state is complete transparency.
Scrolling as reading
During the first four thousand years of Western written language, reading and writing took place on tablets, pieces of animal skin, and rolls of papyrus. The first major break with these technologies occurred with the invention and eventual widespread adoption of the codex, which we are familiar with today as the book. The codex offered several advantages over the scroll: it permitted use of both sides of the page, allowed the joining of smaller texts into a single volume, and facilitated use through pagination. Ultimately, this allowed for easier movement through the text, and more effective forms of organization.
While these advantages soon led to the ubiquity of the codex and the relative disappearance of the scroll, the scroll never vanished entirely—rather it was subsumed within the logic of the codex. In Augustine’s City of God, for instance, the entire book is understood as a singular codex while its interior divisions reflected the quantity of text held on a single roll. Representations of a text in painting and sculpture consistently appeared as scrolls. A similar infolding of previous technologies is visible in digital infrastructures today: as early computer interfaces replaced the traditional office, they also absorbed its accoutrements as icons so that familiarity could engender ease of use.
Most shifts in the reproduction and presentation of text, however, have occurred without any significant alteration of interface. Gutenberg’s revolution progressed without disturbing the form of the book. Thus the transition to digital scrolling is truly unprecedented, as it involves a shift not only in the physical means of reading—as the adoption of the codex did—but also a radical alteration in the production of materials, and, arguably, a change in the way we read.
This shift in reading style has precedents in two other transitions: one from oralized to silent reading, the other from intensive to extensive reading. In late antiquity, reading and writing were extremely technical acts which required specialized knowledge. Early Christians did not easily glean information off a page like modern readers; reading was more like a “decoding” that required two distinct scans: an oral sounding of the continuous string of syllables that revealed words, followed by another oral recitation of the text after words had been identified. Reading here is akin to a software demo, in which spectators gather while a trained specialist handles interactions with the data. The advent of silent reading both condensed and individuated the process—one now could read alone with no witness. The other major shift, from intensive to extensive reading, occurred in the latter half of the 18th century as paper goods became more accessible and printing was standardized. Intensive reading, in which a reader deals with a small canon of established texts, gave way to extensive reading, in which a reader rapidly consumes a wide breadth of materials.
Expanded access to text and, with it, new reading habits helped catalyze revolutions across Europe in the 18th and 19th centuries, as political pamphlets transported information at an unprecedented speed.
If the revolutions of early modernity were kindled with printed pamphlets, our current revolution in social media has been delivered to us conveyor-belt style through an infinite scroll. Introduced on the internet in the late 2000s, infinite scrolling—in which new content is loaded dynamically as the user scrolls downwards—has become the primary method of consumption for almost every social media platform. There is no 'bottom' to speak of on your Facebook, Twitter, or Instagram feed. This technique has attracted criticism hinging on everything from the minor annoyance of losing your place in the endless feed to more serious charges of fueling internet addiction. What kind of content propagates under the logic of the infinite scroll? What are the political ramifications of holding public debate in a constant free-fall? Compact, disposable media has always tended towards the bombastic and hyperbolic, as the political pamphlets and yellow journalism of past centuries will confirm. But what happens when the shelves of the print shop are outfitted with conveyor belts? This continuous scroll of text, a frameless deluge, expands access to raw information, but can equally leave one overly extended, flooded with data and deprived of context. Scrolling succeeds in quantity, but it fails in complexity.
There is also a dark lineage of reading revolutions—those that never truly materialized, yet still dwell at the peripheries of our digital experience.
In 1986, Berkley Softworks released GEOS, an operating system for the Commodore 64 personal computer. Short for Graphic Environment Operating System, it adopted many of the features of its competitors—windows, a desktop, icons—with a key difference. Instead of navigating through the contents of a window by scrolling, GEOS users clicked a dog-eared corner to flip through a metaphorical stack of pages. While GEOS never took off, stacking has continued to lurk in the background as the main alternative to scrolling. In fact, the first hypermedia system, HyperCard, a precursor to the modern internet, was constructed with a card-stacking metaphor central to its design. The system revolved around stacks of virtual cards that could be flipped through, much like a Rolodex. Empowering the user was a central aspect of HyperCard: stacks were user-generated and customizable. Scrolling is a gesture tasked with retrieving more information for the user within a document, while a hyperlink retrieves more information from outside the document. In HyperCard, these two gestures were one and the same. Traversal within documents and meta-traversal of documents employed the exact same atomic gesture of flipping to the next card in a stack. While it anticipated many features of the modern internet, HyperCard was limited to files on a user’s local hard drive. This key limitation led to the project’s cancellation in 2000.
Books, scrolls, clay tablets, and webpages are all essentially means of structuring documents—providing an ordered interface through which we may access their data. Libraries, hypertext, and the internet are means of structuring relationships between documents, so that we might interface with their network. HyperCard was unique in that it offered novel solutions to both problems, recognizing that documents and their networks now shared a single visual plane.
One remnant of stack-based interfaces still lingers in social media, however, which largely function as information retrieval services; you browse Facebook or Twitter to gather novel information about friends, the news, or just to absorb content. As a user, your main power is to navigate this content visually by traveling along a one-dimensional path, the order of which is typically curated by an algorithm. In Facebook this arrives in the form of a newsfeed, which is populated by a linear thread of card-like posts. Tweets similarly adopt a card-form. Of course, both services work to distinguish their platforms from the content that fills them. That is, no one would ever consider Facebook itself to be a post, or Twitter to be a single tweet. Your movement here extends both through scrolling and card-flipping, yet the greater milieu is fixed as something entirely different. But what if, as in HyperCard, traversal within webpages and between websites was accomplished through a single gesture? Can a web be envisioned in which services that aggregate content become archaic? In which the atomic navigational step carries the range of a search engine and the specificity of a scroll?
In the mid-1990s, Douglas Engelbart—credited with the mouse, networked computers, and the window—proposed Open Hyperdocument System (OHS), a “knowledge ecosystem” that would encapsulate the information of the world. Some of its basic elements—a hypermedia system much like HyperCard, in which any point of any piece of media could link to a different point in another, and view control, in which all documents could be displayed according to a flexible choice of options—remain largely unrealized. Similarly, in 1960, Ted Nelson—who coined the terms “hypertext” and “hypermedia”—proposed a “global hypertext publishing system” known as Project Xanadu. While Xanadu also failed to materialize, Nelson has since criticized the modern web as “an imitation of paper” that oversimplifies many of hypertext’s possibilities. Indeed, both Englebart and Nelson have gone on to lament the web’s fixation on linking and scrolling; for both, it fails to live up to the potential of the human mind, which elegantly leaps from concept to concept, fluidly traversing across levels of semantic reasoning and drawing new patterns along the way.
Apart from these faltering glimmers of a different form of digital movement, the prevailing paradigm, largely based on linking to a file and then searching and scrolling through the files, has remained. A gesture which misses the true promise of new media to more closely align with how our brains think and connect ideas. Our minds do not think inside of pages, files, or apps. We think in concepts, we dart around fluidly at whatever level of detail suits the moment, connecting the dots, sparking moments of insight. These strange and failed alternatives remind us of the extent to which present information technology has only served to automate and adapt us to a linear, paper-based world.
ZAK ZIEBELL B/RISD '19 still lurks the forums.