Memory Machines: The Evolution of Hypertext

Memory Machines: The Evolution of Hypertext

by Belinda Barnet


View All Available Formats & Editions
Choose Expedited Shipping at checkout for guaranteed delivery by Thursday, July 25


This book explores the history of hypertext, an influential concept that forms the underlying structure of the World Wide Web and innumerable software applications.

Product Details

ISBN-13: 9781783083442
Publisher: Anthem Press
Publication date: 12/01/2014
Series: Anthem Scholarship in the Digital Age Series
Pages: 192
Product dimensions: 5.90(w) x 8.90(h) x 0.70(d)

About the Author

Belinda Barnet is a lecturer in media and communications at Swinburne University, Melbourne, Australia.


Read an Excerpt

Memory Machines

The Evolution of Hypertext

By Belinda Barnet

Wimbledon Publishing Company

Copyright © 2013 Belinda Barnet
All rights reserved.
ISBN: 978-0-85728-196-8



How does one write the story of a computer system? To trace a technical history, one must first assume that there is a technical 'object' to trace – a system or an artefact that has changed over time. This technical artefact will constitute a series of artefacts, a lineage or a line. At a cursory level, technical 'evolution' seems obvious in the world around us; we can see it in the fact that such lineages exist, that technologies come in generations. Computers, for example, adapt and adopt characteristics over time, 'one suppressing the other as it becomes obsolete' (Guattari 1995, 40). But are we to understand this lineage from a sociological, an archaeological or a zoological perspective? And what is a technical artefact?

I need to address these questions here for two reasons. First, because it is impossible to write a technical history without defining how that history will be constructed, and second, because these questions also concerned Douglas Engelbart, one of the early pioneers whose work we investigate in this book. The relationship between human beings and their tools, and how those tools extend, augment or 'boost' our capacity as a species, is integral to the history of hypertext and the NLS system in particular.

Traditionally, history has ignored the material dimension of technical artefacts. Historians are interested in tracing cultural formations, personalities and institutions, and especially the social 'constructions' they erect around themselves. Technical artefacts don't have their own history; they are perceived as the products of culture. I have always found it strikingly odd, even offensive, that viruses or star formations are allowed to have their own history, but technologies are not. It's as though silicon (or calculus, or mathematics) has no existence outside of its function for human beings. As philosopher Daniel Little writes in a blog post:

History is the sum total of human actions, thoughts, and institutions, arranged in temporal order. Call this 'substantive history'. History is social action in time, performed by a specific population at a time. Individuals act, contribute to social institutions, and contribute to change. People had beliefs and modes of behavior in the past. They did various things. Their activities were embedded within, and in turn constituted, social institutions at a variety of levels. (Little 2011)

I do not have the space to unpack this in more detail here (that is another book for another time, but interested readers may find my article on the topic). Suffice it to say that I think the material dimension of technical artefacts is important, and I pay attention to that here. Technologies like microfilm or digital computing have their own materiality, their own limits and resistances, and these limits affect the process of invention. Technical prototypes in particular are more than just the product of one person's genius; they are physical artefacts, and they generate new devices and new systems by demonstrating what is possible with steel wheel-and-disc integrators or with FORTRAN. Computer languages like Pascal also have their own control structures, procedures and functions that influence the shape of the software. There is actually a historical approach that is interested in how objects change over time, but it does not come from the humanities. It comes from evolutionary biology.

Since the early days of Darwinism, analogies have been drawn between biological evolution and the evolution of technical artefacts. In the background, there has been a long and bloody Hundred Years War among cultural anthropologists as to whether human culture as a whole can be said to evolve (Fracchia and Lewontin 2002, 52). From the middle of the nineteenth century on, and arguably before this, scholars started remarking on the alarming rate at which technological change was accelerating. This sense of 'urgency', that technological change is accelerating and that we need to understand how it occurs, comes through most strongly in Engelbart's work. As he put it in an interview with the author, 'technologies are going to make our world accelerate faster and faster and get more and more complex, and we're not equipped to cope with that complexity [...] [We] need to establish a balanced co-evolution' (Engelbart 1999). This would be the most important project mankind could undertake.

The analogy of technical 'evolution' can only go so far, however. Technological systems are not like biological systems in a number of important ways – most obviously the fact that they are the products of conscious design. Unlike biological organisms, technical artefacts are invented. Silicon does not rise up and make itself into a computer; Pascal or C++ does not coalesce on the screen to form a hypertext system. Consequently, the mode of transfer of 'ensconced information' – designs, techniques and processes – is different from biological organisms.

We must also be clear about what we mean by 'evolution'. 'Change over time' is a good place to start, but it is far too vague to be useful; political parties, ice cream and conversations also change over time, but cannot be said to 'evolve'. On the other hand, it is dangerous (and conceptually vacuous) to import the language of genetics into historiography. As palaeontologist Niles Eldredge puts it:

It is not ipso facto wrong to seek parallels between biological and material cultural evolution in an attempt to clarify the underlying ontological structure and causalities within each system [...] [But] before such comparisons can be made, however, it is important to start with a definition of 'evolution' that is both suitable and appropriate to both systems. (Eldredge 2011, 298)

For our purposes here, Eldredge's definition is perfect: 'the long-term fate of transmissible information' (Eldredge 2011, 298). By 'transmissible information', we mean not just designs and innovations (for example hyperlinks or the mouse), but also the techniques and processes that might be said to make hypertext unique as an information system (I will be arguing that is the organizational technique of association). To explain this more clearly, I spoke to Eldredge about trumpets, trilobites and historiography.

Tracing a Technical Artefact

In both biology and material cultural systems, history is indeed staring you in the face when you look at a wombat or a [technical object]. But there is no way to divine that history unless you compare a series of objects that you assume a priori are related. (Eldredge and Barnet, 2004)

Professor Niles Eldredge collects things for a living, and there are two great collections in his life. The public one is on display at New York's Museum of Natural History; its 1,000 individual specimens stretch floor to ceiling for 30 metres across the Hall of Biodiversity. There are beetles, molluscs, rotifers and fungi, spiders, fish and birds, all arranged into genealogical groups. The other collection is private; it spans an entire wall in his home in rural New Jersey. This collection contains over 500 specimens, but of the 'musical rather than the biological variety' (Walker 2003, 41). He collects cornets, a type of musical instrument. There are silver and gold ones, polished and matte, large and small, modern and primitive. Ever the biologist, Eldredge has them arranged in taxonomic relationships of shape, style and date of manufacture. Much of the variety in cornet design is based on the way the pipe is wound.

Late in 2002, Eldredge's curiosity got the better of him. He decided to feed these specimens through the phylogenetic computer program he uses for his trilobites, to apply the 'scientific method' to technical evolution for the first time. As usual, he asked the computer to come up with all the possible evolutionary trees and then make a 'best guess' based on the existing specimens. The results were astounding. Compared to the phylogenetic diagram for trilobites, the diagram for a technical artefact seemed much more 'retroactive': new designs could borrow ideas from the distant past. Eldredge's musical instruments could step outside time.

In the world of living things, there are basically only two ways creatures can obtain a characteristic: by inheriting it from a previous generation or by evolving it in the present one. This last form of evolution is itself the subject of debate; an organism can't change its DNA in one lifetime. The only proven exception is found in the world of viruses. Biological organisms evolve gradually over thousands of generations, subject to natural selection. If a species dies out – biological 'decimation' – its lineage dies with it. But technical artefacts are different. With technical evolution comes the capacity to co-opt innovations at a whim. Time after time, when one of Eldredge's cornets acquired a useful innovation, other designers quickly copied the idea.

Technical artefacts are not dependent on the previous generation; they can borrow designs and innovations from decades or even centuries ago (retroactivate), or they can borrow from entirely different 'branches' of the evolutionary tree (horizontal transfer). As Eldredge put it:

The key difference [between technical evolution and biological evolution] is that biological systems predominantly have 'vertical' transmission of genetically ensconced information, meaning parents to offspring [...] Not so in material cultural systems, where horizontal transfer is rife – and arguably the more important dynamic. (Eldredge and Barnet, 2004)

The lines in the cornet evolutionary tree were thoroughly confused. Instead of a neat set of diagonal V-shaped branches, a cone of increasing diversity, you would see flat lines from which multiple instruments appeared. When this happens in biology, it implies explosive radiation – the appearance of multiple new species in a geologically short period. In the biological world it is vanishingly rare (though as Eldredge argued with Stephen Jay Gould in their 1972 theory of 'Punctuated Equilibria' (Eldredge and Gould 1972), it is not impossible). The striking thing is that it appears to characterize the evolution of technical artefacts. In the cornet diagram, the gradual passage of time and generations does not precede the development of characteristics. Innovations appear spontaneously, often with no physical precursor.

Most striking of all, and most relevant to the history we are about to trace, outdated or superseded models could reappear with new designs, as if they were held in memory and only needed a certain innovation to burst into activity again. Technical visions from the past can be resurrected. This is what we mean by 'retroactivity'; technical designs can reappear, incorporate new bits and pieces from whatever is around at the time (microfilm, for example, was the newfangled thing in Vannevar Bush's era, and digital computing in Doug Engelbart's era) and then rapidly evolve in a single generation.

[Virtually] all permutations and combinations were possible – including 'retrofitting' older designs with new ideas. Bottom line: information can easily spread back across separate branches (lineages) after they are already established as phylogenetically distinct. (Eldredge 2011, 301)

In biological evolution, when branches diverge, they diverge irrevocably; similarly, when branches die out, they cannot reappear. Technical artefacts are different. There is no extinction; nothing is irrevocable. That is why technical visions, or 'images of potentiality' as we call them in this book, can be influential over decades. In tracing the history of hypertext, then, we will be paying attention to instances of transfer and retroactivity – to innovations jumping from one system to another – but also to images of potentiality that appear to have longevity and recur over time. To abiding dreams.

So we have established a means by which technical artefacts evolve: transfer and retroactivity. But what, exactly, is hypertext, and what is the 'transmissible information' that we are tracing? As Noah Wardrip-Fruin observes in his essay 'Digital Media Archeology' (2011), in approaching any media artefact it is important to understand the systems that support it, and for digital media in particular, the specific 'operations and processes' that make it unique:

Digital media are not simply representations but machines for generating representations. Like model solar systems (which might embody a Copernican or geocentric perspective while still placing the sun and planets in similar locations), the operational and ideological commitments of digital media works and platforms are visible more in the structures that determine their movements than in the tracing of any particular series of states or outputs. (Wardrip-Fruin 2011, 302)

To define the 'operations and processes' behind hypertext as an information system, Ted Nelson is the logical first port of call. His original 1965 definition of hypertext was 'a body of written or pictorial material interconnected in such a complex way that it could not conveniently be presented or represented on paper' (Nelson 1965, 85). This is a useful definition because it emphasizes the most obvious aspect of hypertext: interconnectivity. But it also defines hypertext by what it is not: paper. This does not tell us very much about the underlying processes that might define hypertext as an historical artefact, however.

Nelson's next definition of hypertext is more useful for our purposes. In his 1974 book, Computer Lib/Dream Machines (the copy I refer to here is the 1987 reprint), Nelson proposed a definition of hypertext that emphasized branching or responding text – text that automatically extends itself upon request. 'Hypertext means forms of writing which branch or perform on request; they are best presented on computer display screens' (Nelson 1987, 134). This definition is useful in that it emphasizes that the 'interconnections' identified in the first definition should be automated upon request; they should go somewhere, do something, 'perform' or expand. They are also best read at a screen. This distinguishes hypertext from, say, proto-hypertexts like Cortazar's Hopscotch (a common example in the literature); although these works demonstrate deep interconnectivity, the interconnections are not automated. What Nelson meant by the word 'perform', however, also included some computer-based writing systems that haven't gained mass adoption.

One of those systems was stretch-text, a term Nelson coined in 1967 to mean text that automatically 'extends' itself when required – more like zooming in on a sentence than following a link. In the manner of a link, it still needs to be activated or automated, but it is not based on passing through to another node. The important point about stretch-text is that it doesn't fragment the text; it doesn't take you from one place to another. As George Landow observes, instead of following a link in the text, a stretch-text document 'retains the text on the screen that gives context to an anchor formed by a word or phrase even after it has been activated' (Landow 2006, 95). Although numerous systems have attempted to realize Nelson's conception of stretch-text, it has not gained mass adoption like the 'chunk-style' hypertext Nelson also identified in Computer Lib/Dream Machines; chunk-style hypertext consists of modular units of text or media connected by links (like most websites).

As Wardrip-Fruin wrote in a 2004 paper, although there have been excursions into the other forms of hypertext posited by Nelson (for example, the attempts to build stretch-text), chunk-style hypertext is now the dominant form. The most obvious reason is the Web, the largest chunk-style hypertext system in existence. Another reason might be that all the early hypertext systems (for example, NLS, HES and FRESS) were chunk-style – with the exception of Peter J. Brown's Guide, which was, as Mark Bernstein points out, 'the first commercial hypertext system, and which was deeply stretch-text' (Bernstein 2012).


Excerpted from Memory Machines by Belinda Barnet. Copyright © 2013 Belinda Barnet. Excerpted by permission of Wimbledon Publishing Company.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Table of Contents

Foreword: To Mandelbrot in Heaven – Stuart Moulthrop; Preface; Chapter 1: Technical Evolution; Chapter 2: Memex as an Image of Potentiality; Chapter 3: Augmenting the Intellect: NLS; Chapter 4. The Magical Place of Literary Memory: Xanadu; Chapter 5: Seeing and Making Connections: HES and FRESS; Chapter 6: Machine-Enhanced (Re)minding: The Development of Storyspace; Conclusion; Notes; Bibliography; Index

What People are Saying About This

From the Publisher

‘Belinda Barnet has given the world a fine-grain, blow-by-blow report of how hypertext happened, how we blundered to the World Wide Web, and what other things electronic literature might still become.’ —Ted Nelson, hypertext pioneer

‘This is a fine and important book, the first to capture the rich history of ideas and people that led to the World Wide Web. “Memory Machines” carefully examines what the key figures were trying to do and judiciously explores what they accomplished and how the systems we now use daily sometimes exceed their dreams and sometimes fall embarrassingly short of their early achievements.’ —Mark Bernstein, Chief Scientist, Eastgate Systems

‘This is well-researched and entertaining story, full of personal anecdotes and memories from the people who built these important early systems.’ —Professor Dame Wendy Hall FREng, Dean of Physical and Applied Sciences at the University of Southampton

‘Walter Benjamin wrote that “It is not that what is past casts its light on what is present, or what is present its light on what is past; rather...what has been comes together in a flash with the now to form a constellation. “Memory Machines” is, even for one among its participants, such a constellation of the now.’ —Michael Joyce, Professor of English at Vassar College, New York

Customer Reviews

Most Helpful Customer Reviews

See All Customer Reviews