May 18, 2004
What Hypertext Is
I’m working on a short paper for ACM Hypertext 2004, which will be at UC Santa Cruz this August (and where Matt Webb and I will be offering a tutorial on blogging). The short papers deadline is the 28th of this month. The working title of my paper is “What Hypertext Is” and my goal is to provide a 2-page answer to the old chestnut “What is Hypertext?” I want to give a much, much better answer than you find in many places — such as the current everything2 entry, which begins: “Hypertext is nothing more than the inclusion of links within a body of text.”
I’m including a draft below, and would definitely appreciate comments. I can’t make it any longer, but I could substitute, clarify, reconsider, etc. Here’s a preview:
We can now, based on our examination of Nelson’s texts, provide a relatively concise definition of hypertext appropriate for a world familiar with the Web: “Hypertext is a term coined by Ted Nelson for textually-focused forms of hypermedia (new media that branch or perform on request). Examples include the link-based ‘discrete hypertext’ (of which the Web is one example) and the level-of-detail-based ‘stretchtext.'”
What Hypertext Is
Noah Wardrip-Fruin
ABSTRACT
Over the past couple decades, as the term “hypertext” has gained a certain popular currency, a question has been raised repeatedly: “What is hypertext?” This paper serves to provide a relatively-concise answer to this question. Hypertext is a term coined by a 20th Century thinker — and it is in this way similar to “natural selection” or “communism.” In this case the thinker was Theodor Holm (“Ted”) Nelson, and in this paper two of his early publications of “hypertext” are used to determine the term’s meaning: the 1965 "A File Structure for the Complex, the Changing, and the Indeterminate" and the 1970 "No More Teachers’ Dirty Looks." It is concluded that hypertext is a term for textually-focused forms of hypermedia — which are media that “branch or perform on request.” Finally, the understanding of the term hypertext presented here is considered in light of the work of Vannevar Bush and Douglas Engelbart, recent critiques of hypertext, and possible future directions for the hypertext conference and community.
INTRODUCTION
When we ask the question, “What is hypertext?” we are asking a question fundamentally similar to questions such as “What is psychoanalysis?” “What is natural selection?” and “What is communism?” We are asking about the meaning of a term coined by a particular 20th Century thinker. While such terms are often taken in a variety of different directions after they are coined (much as Stalin took usage of the term “communism” in a direction that Marx is unlikely to have imagined) it is generally agreed that serious discussion of the meanings of such terms must begin with the work of thinker who coined them.
In the case of “hypertext” the term was coined by Theodor Holm (“Ted”) Nelson. This paper examines two of Nelson’s early publications of the term in order to provide a definition. It is found that definitions of hypertext that focus on “the link” or “structured knowledge work” are not supported by an examination of Nelson’s work, whereas definitions that focus on media are supported. Implications for recent hypertext critiques and possible futures for of the hypertext conference and community are then presented.
HYPERTEXT, FILM, AND MEDIA
While Nelson may have presented the terms earlier, his first significant publications of the terms “hypertext,” “hyperfilm,” and “hypermedia” occur simultaneously — in the 1965 paper “A File Structure for the Complex, the Changing, and the Indeterminate” [1]. As the title suggests, this paper was primarily concerned with outlining a file structure (the Evolutionary List File, or “ELF”) inspired by Vannevar Bush and intended for personal use by knowledge workers. However, a final section (titled “Philosophy”) makes a departure, noting that file structures like the ELF “make possible the creation of complex and significant new media, the hypertext and hyperfilm.” Two paragraphs later, Nelson expands the term “hypertext” in this famous sentence: “Let me introduce the word ‘hypertext’ to mean a body of written or pictoral material interconnected in such a complex way that it could not conveniently be presented or represented on paper.” In the following paragraph “hyperfilm” is briefly mentioned again: “The hyperfilm — a browsable or vari-sequenced movie — is only one of the possible hypermedia that require our attention.”
It is worthwhile to note the following: (1) “hypertext” and “hyperfilm” are coined within the same sentence; (2) both hypertext and hyperfilm are characterized as “new media”; (3) the larger category in which at least the hyperfilm is included is “hypermedia”; and (4) while the material Nelson offers in this brief section does not explicitly contradict definitions of hypertext that focus on the link, links are not mentioned.
FORMS OF ‘HYPER-MEDIA’
Just as Nelson’s 1965 paper was initially concerned with the presentation of a file structure (rather than an explication of “hypertext”), his 1970 “No More Teachers’ Dirty Looks” [2] begins with a critique of concepts of “computer-assisted instruction” founded on a drill-and-practice model. Nelson’s proposed alternative to such systems is “responding resources.” He writes, “Responding resources are of two types: facilities and hyper-media.” On-screen calculators and graph plotters are given as examples of facilities. Nelson then writes of hyper-media:
Hyper-media are branching or performing presentations which respond to user actions, systems of prearranged words and pictures (for example) which may be explored freely or queried in stylized ways. They will not be “programmed,” but rather designed, written, drawn and edited, by authors, artists, designers and editors. (To call them “programmed” would suggest spurious technicality. Computer systems to present them will be “programmed.”) Like ordinary prose and pictures, they will be media; and because they are in some sense “multi-dimensional,” we may call them hyper-media, following mathematical use of the term “hyper-.”
Nelson then presents examples of types of hyper-media that could be made available to students. The first of these is under the heading “Discrete Hypertexts.” Nelson writes: “‘Hypertext’ means forms of writing which branch or perform on request; they are best presented on computer display screens… Discrete, or chunk style, hypertexts consist of separate pieces of text connected by links.” This is the first appearance of the term “link” in the essay.
Nelson next presents a type of hypermedia called the “hypergram” (“a performing or branching picture”) followed by another form of hypertext — “stretchtext.” Nelson writes: “This form of hypertext is easy to use without getting lost… There are a screen and two throttles. The first throttle moves the text forward and backward, up and down on the screen. The second throttle causes changes in the writing itself: throttling toward you causes the text to become longer by minute degrees.” Note that Nelson referred to hypertext as “forms of writing which branch or perform on request.” Discrete hypertext uses links to branch on request. Stretchtext uses no links — instead making a non-branching performance on request. Definitions of hypertext that treat the link or branching as fundamental are therefore clearly insufficient.
Nelson’s essay continues, outlining further types of hypermedia such as the “hypermap” (a smooth zooming interface) and “queriable illustrations” (a type of hypergram). The fact that both “discrete hypertext” and “stretchtext” are situated within this list of examples of types of hypermedia leaves little doubt that hypertext is a subcategory of hypermedia (which, again, are “branching or performing presentations which respond to user actions”). While the behaviors of discrete hypertext and stretchtext are quite different, what unites them is that they are “forms of writing.” While the examples shown of hypergrams, hypermaps, and queriable illustrations all include text, discrete hypertext and stretchtext are primarily textual. Hypermedia, also, are differentiated from “facilities.” They are not tools, but media — “designed, written, drawn and edited, by authors, artists, designers and editors.” From this we can conclude that tools such as spreadsheets and word processors are, along with calculators and graph plotters, not hypertext.
We can now, based on our examination of Nelson’s texts, provide a relatively concise definition of hypertext appropriate for a world familiar with the Web: “Hypertext is a term coined by Ted Nelson for textually-focused forms of hypermedia (new media that branch or perform on request). Examples include the link-based ‘discrete hypertext’ (of which the Web is one example) and the level-of-detail-based ‘stretchtext.'”
IMPLICATIONS
If we provisionally accept a definition of hypertext such as the above, what are the implications?
First, we can understand the relationship that the work of figures such as Vannevar Bush and Douglas Engelbart has to the term “hypertext.” Both, rather than producing work that provides a definition of hypertext, envisioned particular forms of hypertext. Bush’s Memex, with its use of links and trails, can be seen as a form of discrete hypertext when employed as media, and as a hypermedia authoring system when used to record the user’s material. Engelbart’s NLS — with its links and powerful hierarchical organization — can be usefully described as combining elements of the discrete and stretchtext approaches, and again as both a presentation and authoring system for hypermedia. Like Nelson’s proposed Xanadu, the Memex and NLS can also function as what Nelson calls facilities, as tools for knowledge work rather than media. This is similar to the distinction drawn by Nelson in 1965 between the use of his ELF concept for media (enabling the hypertext and hyperfilm) or knowledge work.
Second, a number of “hypertext critiques” collapse. In the new media theory community a number of writers have constructed a false opposition between hypertext and new media. In fact, Nelson’s writing from 1965 makes clear that hypertext has been positioned among “new media” for four decades. Similarly, those who critique hypertext as simplistic, who usually define hypertext using a simple link-based model, are shown to be operating in error. (For example, take Espen Aarseth’s statement, in his 1994 “Nonlinearity and Literary Theory,” that “[h]ypertext, for all its packaging and theories, is an amazingly simple concept. It is merely a direct connection from one position in a text to another.”)
Third, we can imagine certain future directions for the community that calls its work hypertext (from literary work to system building to theorizing related to this range of activities) and for the ACM Hypertext conference. Peter J. Nürnberg’s closing keynote for Hypertext 04 conference (“What is Hypertext?”) suggested that the hypertext community is at a moment that calls for self-definition. At times of crisis within intellectual communities it is common to urge a return to the work of the founding figure. I follow in this tradition, by suggesting that this view of hypertext opens a direction for the hypertext community to focus on types of new media for which text is central. This places us in close proximity with the digital media community, the AI community (particularly NLP), and the electronic publishing community (especially in relation to document engineering and the Web). This is to say, such a definition embraces what we currently create and study, but broadens our self-conception to include, for example, the most interesting language-oriented innovations in computer games, electronic art, educational systems, and other forms of digital media. It takes away the cause of saying, “this isn’t really a hypertext paper” for any textually-focused type of hypermedia.
REFERENCES
[1] Nelson, Theodor H. A File Structure for the Complex, the Changing, and the Indeterminate. Association for Computing Machinery: Proceedings of the 20th National Conference (Cleveland, Ohio, U.S.A.) 1965, 84–100.
[2] Nelson, Theodor H. No More Teachers’ Dirty Looks. Computer Decisions 9, 8 (Sep. 1970) 16–23.
May 20th, 2004 at 2:19 pm
Noah,
Since you are inviting comments on your hypertext definition, here are my two cents worth. I really appreciate your going back to the sources and trying to elucidate Ted Nelson’s concept. I believe however that words and concepts have a life of their own, and if the term has evolved beyond it’s creator’s definition, we should accept this. After all, originally, the term “computer” refered to some manual contraption used by accountants. Returning to the source also assumes that there are no contradictions in Nelson’s writings about hypertext but judging by your quotes there are.
First you quote: “Let me introduce ‘hypertext’ to mean a body of written or pictorial material interconnected in such a complex way that it could not conveniently be presented or represented on paper.” Here, hypertext is a general category that includes as subspecies hyperfilm and something language-based. We could also have hypermusic or hypermulti-media.
Later on, however, Nelson defines hypertext as a subcategory of hypermedia—”performing presentations which respond to user actions.” Hypertext, then, is text-based hypermedia. There are two forms: link-based and “stretch text.” Link based is what most people understand as hypertext, while examples of stretch text would be some Flash movie that uses only language.
Is it really useful to have a category that includes both linked-based and stretch forms of “text,” as opposed to another category (I don’t know what the name would be) that would cover both linked-based and stretch visual artifacts (for instance a Flash movie where the user’s mousing causes objects to change shape), and yet another category for linked-based and stretch multi-media artifacts? What do we do, in such a system, with a Director movie that uses both “goto” buttons, and mouseovers that alter the display, and blends text with music and pictures? And what do we do with a text that stretches by itself, as in a non-interactive Flash movie?
I personally would support a taxonomy that regards hypertext as a particular form of hypermedia, as defined by Nelson. The term would refer to discrete pieces of data, whether visual or textual or aural, connected by voluntarily activated links that bring other such chunks of data to the screen. Hypermedia artifacts that are not hypertexts would be cybertexts. This would include IF, computer games, non-interactive and purely reactive Flash or Director texts, and code-generated poetry. (Then however we would have to take “responding to user actions” out of the definition of hypermedia, and replace it with “performing in a digital environment.”)
And of course there would be hybrids of hypertext/cybertext, such as my example of the Director movie that uses both “goto” buttons, and mouseovers that alter the display.
May 20th, 2004 at 4:16 pm
Marie-Laure, thanks for coming by and commenting.
I guess I was unclear about what Nelson wrote in 1965. He did not define hyperfilms as a subset of hypertext. He defined hyperfilms as a subset of hypermedia. However, this doesn’t invalidate your point, that Nelson says “written or pictorial” material when talking about hypertext in 1965. I can only speculate that the hyperfilm is differentiated from hypertext in that its logic is more filmic than textual. What do you think? Should I change my capsule definition to say that hypertext is hypermedia that operates textually, that operates via a textual logic?
I have to disagree with the “here and now” approach to intellectual history. The original meaning of hypertext is important, and should be borne in mind by careful scholars, even if popular use has shifted. (And “hypertext” is a term for a concept, it is an intellectual term, rather than a term naming an object like a computer.) Otherwise we’re making an argument like, “No need to be aware of what Marx meant by Communism. Wake up, it’s the 1950s, Stalin’s definition is the only one that matters.” I agree with you, of course, that meaning shifts – and meaning will keep shifting – but our job as scholars is to understand that history, rather than be short-sightedly caught in the meaning of the moment. How would we feel about an entry in an encyclopedia of biology, meant to define “natural selection,” that made no reference to Darwin, that didn’t take Darwin’s work as the starting point?
What do you think of another argument, which I could probably find room for in my paper — if we pay attention to what hypertext actually means, then we don’t need the term “cybertext”? Would you agree? Could we say that cybertext is just another term for textual media that “branches or performs on request”? Then we could have subsets of hyper/cybertext like “discrete hypertext” (which is link-based) or “interactive fiction” (which has a world model, a parser, and…). Or would “cybertext” remain useful as a name for a subcategory of hypertext, in which feedback loops are particularly important? I think everything you name in your comment is a type of hypermedia, and if the focus/logic were textual I’d say they’re all types of hypertext.
May 20th, 2004 at 9:29 pm
Hello Noah,
This is not a criticism, and indeed may not help you at all but is in the interest in identifying the bounds of the term and how it can be applied.
Ted said:
“Let me introduce the word ‘hypertext’ to mean a body of written or pictorial material interconnected in such a complex way’
Ted said:
“‘Hypertext’ means forms of writing which branch or perform on request’
Noah said:
“Hypertext is a term coined by Ted Nelson for textually-focused forms of hypermedia (new media that branch or perform on request). Examples include the link-based ‘discrete hypertext’ (of which the Web is one example) and the level-of-detail-based ‘stretchtext.'”
The only item that I feel was relevant when the terms were introduced and now is only implied but not explicitly stated is: whether the ‘body’ and ‘forms’ refer to a type of work in isolation and the phenomenon of multiple works. In other words –- what are the bounds of the ‘form’ and ‘work’? I know the categorisation you have adopted here is to describe new media works, and works that are ‘textual’ in general. This means they can be applied as a blanket term to differentiate from non-textual based and indeed non-new media. How does it apply to works that fit into these but are experienced over multiple works? For instance: alternate reality gaming (immersive and pervasive), if we assume the works are all textual for this point, the user attends to many websites. Would they be considered many hypertexts or one?
Given the systems conceived by Vannevar Bush and Ted Nelson are intended to be a navigation across many of these works (the digital encyclopedia to assist in research and pedagogy) and now this phenomenon does indeed occur in fiction and education, does the term ‘hypertext’ encapsulate this?
Also, one slip here: Peter J. Nürnberg’s closing keynote was in 03?!
May 21st, 2004 at 3:28 am
I don’t know, does the I Ching branch at request? Matter of definition, really… And.. do you really think Nelson would have thought of interactive fiction as hypertext? Do you, Nick?
I think it’s useful to have a paper lay out various of Nelson’s original definitions of the terms, and I like your call to action at the end of the paper – this particularly makes sense in the context of last year’s closing keynote. How about moving a hint of this into the abstract, so people immediately see that this isn’t just about abstract definition but about shaping and perhaps saving a field?
Here’s stuff I noted reading your text:
I didn’t hear Nürnberg’s keynote, but I do wonder whether most of the Hypertext community mightn’t already agree with you – and it’s actually people outside of the Hypertext community you need to be reaching, really. On the other hand, you’re quite right, the “but it’s not really a HYPERTEXT paper” is common, so perhaps this isn’t true. I do know that when I attended my first Hypertext conference I thought hypertext was discrete hypertext and nothing else, and it was only through hearing Frank Shipman and many others talk about spatial hypertext I realised that there are many possible kinds of hypertext. So for me it was *through* the Hypertext conferences I broadened my concept of hypertext. Though not entirely in the direction you wish to take Hypertext.
Speaking of spatial hypertext, how would that fit within your definition? It doesn’t (necessarily) branch or perform on request. But for large portions of the Hypertext community, it’s definitely hypertext. Does anything Nelson writes seem to cover spatial hypertext?
In sum, I think the concise clarification of Nelson’s original definitions of the term is really useful, also as something to build further discussion upon, and I love how it’s a position statemnet and an invitation to discussion – I think that’s a great way of using the short paper format. I like your Implications section best, though obviously you need the first section, but perhaps you could hint more as to the implications and the this-is-why-it’s-important earlier in the paper.
May 21st, 2004 at 9:51 am
To Christy:
You’re absolutely right about when the keynote was, of course. Thanks for catching that.
I’ll have to mull further your question about media like alternate reality games distributed across multiple sites on hypertext networks.
To Jill:
I admit that the IF comment was a bit of a provocation, but I actually think it fits, though perhaps as a borderline case. Which makes me realize another way that the definition I offer should probably be refined. Nelson’s 1970 essay is distinguishing hypertext from AI dreams of computer-assisted instruction. He places a lot of emphasis on hypertext being human-authored, rather than magically produced by the machine through its understanding of human learning. IF was then, and continues to be, carefully human-authored work that branches and performs on request. So, yes, I think if we could go back and ask the 1970 Nelson about IF he’d probably say, “I don’t think it’s a good medium for education, but I think it’s hypertext.” That said, I think Facade is an example of using AI-derived tools for the purposes of creating human-authored (rather than magically machine-reasoned) media. Another interesting (perhaps borderline) area — which I think should have a home at the Hypertext conference.
The I Ching doesn’t branch or perform on request. My understanding is that it requests that humans perform actions and then branch over its text. The I Ching doesn’t do anything itself, however much you request it to.
I think you’re absolutely right that I should preview some of the material from the last section in the abstract. Maybe I’ll cut down the introduction to do that (and, yes, I should take more care with my dates — thinkers whose ideas shaped the 20th Century is not the same as “20th Century thinkers”).
About last year’s closing keynote, I’m afraid it didn’t show a hypertext community that largely agreed with me. The idea that hypertext might be a form of media was, as I remember it, pretty peripheral. And Ted Nelson’s work went without mention, even though Nelson was in the room. Looking over Matt Webb’s notes seems to confirm this, along with my memory of what the keynote argued hypertext is: “structured knowledge work.” During the questions, when hypertext poet Jim Rosenberg raised his hand, I thought it would be to offer the alternative, media-focused perspective — but instead it was a question that fell within the framework the talk had introduced. Of course, I didn’t ask a question myself at the time. In a sense, this paper is my response.
Unfortunately, I’m on the road, and have to leave my internet connection now in order to reach this event on time:
Symposium: Interaction, Immersion and the Illusion of Control.
May 21st, 2004 at 12:52 pm
My problem with sticking to Nelson is that if we define hypermedia as media that “branch or perform on request,” an excellent formula, then hypertext, as the purely textual (language-based) forms of hypermedia, is not a very interesting concept because its definition derives too automatically from the definition of hypermedia. It’s hypermedia that deserves all the attention.
By Ted’s definition of hypermedia, IF and AI based interactive drama and even computer games certainly qualify, though they are not link-based. The branching occurs when users solve problems and take actions, not when they click on a link. I would not call these forms hypertexts, though some people do. For instance Angela Ndalianis, in her new book “Neo-Baroque Aesthetics and Contemporary Entertainment,” calls computer games hypertexts because they have a branching structure.
Now to return to hypermedia, does Ted’s definition imply a computer? Seems to me that only code-based texts can really be said to perform—in the sense that one performs a musical score. (The code is the score, the machine the instrument.) Having a branching structure and being a performing text and performing on request are separable properties. For instance:
Branching print texts such as Oulipo texts do not perform by themselves. It’s the reader who turns the pages, as Noah observes.
Non-interactive Flash movies perform but do not branch.
A code-driven text that branches according to the value of a random number (i.e. by reading the computer clock) performs and branches but not on (human) request. Same for a text that branches when the user mouses over invisible hot spots.
So we need a broader concept that Nelson’s hypermedia to cover all this. And I still think that it is useful to have a term for the kind of branching that is triggered by links. For me that’s hypertext. For Noah, it may be linked-based hypermedia, with linked-based hypertext for the language-based version.
The bottom line is that there is no “true” definition, only more or less useful ones.
May 21st, 2004 at 6:51 pm
This is an interesting debate but I’m not sure it’s necessarily a useful one. While it seems to me that your interpretation of Nelson’s definition is correct and that, at this point, it might be useful to the hypertext (literature) community to expand the breadth of its focus, it also seems to me that there would be problems in losing the distinction between “hypertext fiction,” which has more or less been defined as a genre, and “interactive fiction,” which has also been more or less defined as a genre.
At this point, hypertext already has about 4 or 5 definitions. No matter what Nelson said or for that matter what is debated at the Hypertext conference, the term is unlikely to fold back neatly into its (larger) original box. Some people writing about hypertext will understand it to mean “html documents on the World Wide Web,” some will understand it to mean “documents or literary works that use links that branch or perform on request,” some will mean “hypertext fiction as opposed ‘interactive’ fiction or flash poetry,” some will use it to describe a concept, others a technique, others a genre. I haven’t followed its progress in the OED, but by now we should have several numbers under that original definition of hypertext. I guess my question for you, Noah, is why should we bother retro-fitting the term to its more expansive original definition? In what way, other than as a matter of debate at the Hypertext conference (or perhaps to expand the range of papers and creative work heard and shown at that conference) is it useful to do so? “Electronic Literature” and/or “New Media Literature” suffice as general terms, no? And if we correct our “hypertext” usage to mean all electronic texts that perform on request, what term do we use to distinguish between “hypertext fiction of the link and node variety” and all the other forms that hypertext could encompass? I agree with Marie-Laure. The history of “hypertext” is interesting, and if there are “hypertext snobs” clamoring to exclude other forms of hypermedia from some imaginary canon, they could certainly be corrected with Nelson’s original definition(s). But beyond that, why bother?
May 21st, 2004 at 7:51 pm
Noah, just a few notes in reply since you give me the honorable task of representing the hypertext critics [more here]:
On one side we have a narrow definition of hypertext — a direct connection from one position in a text to another — on the other side we have a broad definition —new media that branch or perform on request.
If we examine the second definition, we see that branching (like stretching) is a kind of performance (agreed?), and so the definition might as well read: new media that perform on request.
And then, we see that we can simplify even further, since all new media perform on request, right?
So: hypertext/-media = new media.
If we keep the narrow definition, hypertext retains its focus; if we adopt the broad definition, hypertext loses its terminological power, and is in danger of becoming what Nelson once aptly called “cybercrud”:
“The mere fact that a computer is involved in something has no bearing on is character or validity” (Computer Lib, p.8)
Come to think of it, even my TV performs on request when I switch channels (or is that mere branching?) –Stupid old TV, pretending to be new media!
The dilemma of the hypertext research community is that on the one hand they need an identity that is sharper than just “new media”; but, on the other hand, if the idea becomes too narrow then they lose out on cool new stuff. So they seem to be caught between the (by now) mundane link-based hypertext and the new stuff that could be anything (games, AI, “wheels for the mind” etc.)
BTW: doesn’t Nelson’s insistence that hypermedia will not be “programmed,” rather exclude things like computer games from the neo-orthodox definition?
May 22nd, 2004 at 8:11 am
Marie-Laure, Scott, and Espen — thanks for your additional comments. This is getting to be a substantive (and helpful, for me) conversation.
First, let me apologize that my time is again limited this morning. I’m still in NYC, the symposium is ongoing, and so on. I’m not sure how much I’m going to be able to respond to before I need to leave.
Let me move backwards through the comments. Espen, I agree(d) with you that the less-than-careful readings of some hypertext theorists of 10-15 years ago require(d) correction. No argument there. I’m going to go forward here assuming some of your comments are jokes (e.g., we all understand the distinction between game design and game programming, even if the same person may be involved in both; we all understand that a TV-like device, with many streams of video and a way to move between them, if constructed as a single piece of media, would be hypermedia — Lorna after all was operated via a laserdisk remote).
No, “hypermedia” and “new media” are not synonyms. I consider new media to be using the computer as an expressive medium. It’s very large. It includes, as Marie-Laure has pointed out, things that function without any meaningful input from the audience. So, for example, Dakota is new media, but it is not hypermedia or hypertext. Less formally, hypermedia places its weight on human authorship. We have fewer dreams of “AI authorship” (or whatever) than we used to — but the emphasis on human authorship still makes some things new media that wouldn’t be hypermedia.
I think the reason it’s important to retain “branch or perform” (at least for the purposes of my paper) is that people are familiar with link-based branching, but not with non-link-based performance, as part of hypertext’s definition.
Ah – and I just got a phone call from the person I’m meeting this morning, and have to leave. More as soon as I can!
May 22nd, 2004 at 11:34 am
Noah, I guess it all comes down to what performing on request might mean. It is awfully vague. Clearly, a multi-channel TV is a medium that “performs on request”. If you want to exclude TV channelsurfing from hypermedia then you probably need a more precise definition, which should focus on “work” as opposed to medium.
As for games and “the distinction between game design and game programming” you are not saying that games are hypermedia when they are being designed and not when they are being programmed, are you? Or that games made with game dev tools that don’t require programming are hypermedia? Such a distinction is possible, but not really fruitful.
Programming is another very wide term, and it could be (and probably has been) argued that even setting up guard fields in Storyspace is a form of programming. Not to mention scriping games, or balancing variables as part of tuning.
I (wrongly?) understood your definition to address works of hypermedia rather than the non-programming part of the process of constructing them, or the tools with which they are built.
The problem with distinguishing between computerized texts and “humanly-performed” texts such as the I Ching or various OuLiPo texts, is that it is not very hard to make a computer play the human part. And vice versa, as the choose-your-own-adventure books show. And then the distinction between “old” and “new media” becomes revealed as ideological. To claim that “new media” have to be digital is also to exclude a priori the possibility of non-digital new media, no?
May 22nd, 2004 at 10:47 pm
Okay, picking up some previous threads…
To Christy:
If I understand what you’re asking correctly, the question is similar to asking whether a fiction told as entries in an electronic calendar should be viewed as having the responsive characteristics of the calendar program (which might be many) or of the medium for the entries (say, plain text). I sense this is probably something that media theorists have argued about. I wonder if anyone out there has any references to share with us on this…?
To Jill:
What you ask about spatial hypertext is a good question. Here’s my first-thought answer. I think most of what people talk about as spatial hypertext is more in the tools direction, rather than the media direction, which makes it more like a “facility” in Nelson’s terms. But there is certainly spatial hypertext used for media — such as in Jim Rosenberg’s work. And in these cases there’s the spatial navigation and selection that seems like a textual version of some of what Nelson describes hypergrams doing. So, my preliminary answer is “yes.”
To Marie-Laure:
I think what you say is fine until the end. Would you say there are only “more or less useful” definitions — not more or less accurate ones — of communism, natural selection, and psychoanalysis? When we’re dealing with terms coined by particular thinkers I think there are more and less accurate definitions, more and less intellectually responsible definitions. I think we must start with, or at least explicitly include, the definitions offered by those who coined the terms.
To Scott:
Scott, I think you and I just start with different assumptions. It seems that your assumption is that our field’s historical amnesia will continue indefinitely. My assumption is that the amnesia is temporary. I assume that, a decade or two from now, our field will value its intellectual history as much as other fields of intellectual endeavor do. The New Media Reader and this paper are just to help the process along. (And I assume we will call link-based hypertext something like, “link-based hypertext.” Or use Nelson’s terms: “discrete hypertext” and “chunk-style hypertext.”)
To Espen:
Espen, my interpretation of what Nelson’s saying is more straightforward, I think, than any of the options you propose. He’s saying that calling the process of authoring hypermedia “programming” puts the emphasis in the wrong place. Maybe it’s correct to call the process of creating a 3D shader, or a web server, programming. He thinks of that kind of work, however, as the creation of the system for presentation. He wants to emphasize that the work done by authors and designers, creating the content that is presented, shouldn’t be called programming. It might seem obvious to us now, but in 1970 it might not have been. (And, certainly, there are some situations for which it’s a problematic distinction — such as those in which people design and code as one process. But this is not something Nelson raises repeatedly, as he does with “branch or perform” and the emphasis on human authorship.)
As for the I Ching, I don’t think I follow you. It would also be easy to have a human or a computer be your opponent in Pong. I don’t think that reveals the act of drawing the distinction as ideological. It does make a difference whether an act is performed by a computer or a human. A printed version of an email narrative, with pages you turn, is different from receiving the same text as email. (Am I totally missing your point here?)
May 23rd, 2004 at 3:54 pm
Noah, thanks for elaborating on what Nelson might have meant by programming. It seems to me then, that games, which are so obviously programmed even in the way they are balanced for gameplay, do not fit the category of hypermedia very well. (Not that I thought so.) Even non-computer games, (or, rather, their players) are programmed. Games are systems, after all.
It is interesting that you say “It does make a difference whether an act is performed by a computer or a human”, when it seems to me that Nelson is saying the exact opposite in the quote from Computer Lib above. Your email example is too generic; the difference depends on a lot perceptual and cognitive aspects that has little to do with “print” or “computer”; such as paper/print quality, lighting conditions, screen format and quality (or perhaps synth voice gen) the email software, hardware, the print font, etc. The perceived difference is also highly personal, and contextual. As long as the reading conditions are satisfactory, I don’t think the difference in your example makes any difference at all.
I can’t speak for the field, of course, but it seems to me that one of the most important realisations in game studies in the last few years is that it makes no sense to look at computer games in isolation from games in general. Yes, one can define computer games tautologically as games that are “using the computer as an expressive medium”, but that makes for an extremely arbitrary and analytically useless category, just like, dare I say it, “digital narrative, poetry, and art.” There may be sound ideological, economical and political reasons to emphasise the digital, but no scholarly reasons, as far as I can see.
May 23rd, 2004 at 11:21 pm
Espen, I think we may have to agree to disagree about when it makes a difference to use a computer.
It sounds as though, in the first paragraph above, you’re saying that something like designing the way an object moves (or system operates) is programming when it’s done for a computer game. But if the same task were being done with non-computer tools we certainly wouldn’t call it programming. We’d probably call it design. And my interpretation of Nelson is that he calls on us to call it design even when a computer is involved.
My reading of your next couple paragraphs is that, for works of media, context makes no difference. Only formal characteristics of a work matter. As a writer/artist I just can’t follow you there. The Impermanence Agent was created specifically for the context of the personal computer, web browser, and 1990s network environment. Having all the same functions in some other context (a non-computer context, a different computer context) would not result in the same experience. For me, saying it’s analytically useless to have a category of “digital narrative” is like saying it’s analytically useless to have a category of “street theatre.”
So, did I misunderstand what you meant, or are we simply not going to see eye-to-eye here, or is there a third option?
May 24th, 2004 at 2:43 pm
Well, Noah; you misread me if you think I said context makes no difference. Quite the opposite, so look again, please: the difference is contextual, I said. All sorts of things can make a difference, I said. But the mere fact that a CPU is involved need not be one of the differences that actually makes a difference. Here, I agree entirely with Nelson and his cybercrud critique, but you don’t?
As for programming vs designing, for me they are pretty much synonyms, and digital/analog makes no difference in that regard. There were progams long before digital computers (the word itself is from Greek), so it would be ahistorical to confine its meaning to computerbased work. (The original meaning of a concept is important, yes?)
And as for “street theatre,” as far as I know it is a historical and not an analytical term at all, so yes, it is probably quite useless as such. Those who want to make the case that “digital narrative” is a useful analytical category will have to define it formally. Personally, I don’t think a useful or even interesting definition (distinguishing between digital and non-digital narrative) can be made. But that should not stop anyone from trying; so go ahead and prove me wrong, by all means.
May 24th, 2004 at 4:02 pm
A useful discussion. Thanks to all.
On “the field’s collective amnesia”: I think the amnesia is not much of a problem, today, amongst scholars and serious students. Sure, for some people, hypertext is a Microsoft product that their little nephew calls MSIE. The NMR is obviously an important step in correcting whatever amnesia might have been caused.
On the question of papers being “not really hypertext papers”: this rarely hinges on the question of “what is hypertext”, but is more often asking whether the chief point of interest in the paper is hypertext or something else. If, for example, you’re addressing a question that is mildly interesting to hypertext theory and absolutely fascinating to research in (say) internet protocols, it makes sense to ask “Is this really a hypertext paper” as a shorthand.
I’m not convinced the textual focus is essential to the definition. It’s certainly never been applied against audio hypertext (Aarons), cinematic hypertext (HyperCafe, which one the first Engelbart prize), or visual hypertexts such as Inigo Gets Out.
I don’t think I’ve ever heard anyone argue that spatial hypertext isn’t hypertext.
I do think it’s useful that the definition does not embrace everything on the computer screen.
May 25th, 2004 at 11:04 am
To Espen:
If designing and programming are pretty much synonyms for you, then I understand why we read Nelson differently when he argues for calling the design of hypermedia “design” rather than “programming.” (I don’t think he would make the argument if he thought they were understood as synonyms.)
As for the other issue, it may simply be that I don’t understand what you mean by analytic categories. I take it that you view context as useful, but not useful for the construction of analytic categories. I’m pretty sure people still do street theatre, so calling that a historical category doesn’t seem quite right to me. (I’m also not convinced that the email narrative example isn’t useful.) Are there other options for what we might call context-based categories?
To Mark:
I’m glad to have you join us, especially as I know you share my belief that attention to Nelson’s work is appropriate for our community.
I guess what I see as the field’s collective amnesia is present when people argue for definitions of hypertext that are link-based (many problems with this, starting with the fact that stretchtext is clearly a kind of hypertext and isn’t link-based). These link-based definitions are still the most prevalent definitions in the literary community (and even appear in places like everything2).
On a more mundane note (and I guess blogs are good for mixing types of discussion) I recently received some off-blog advice that I should cut the text of the paper significantly in order to make room for at least four more references. The person giving this advice said that HT program committees don’t tend to look kindly on papers with as few references as this one. I’m a bit resistant, because I feel that the Nelson references are the ones I’m discussing and space is already in short supply for the things I’d like a 2-page paper to do. I know you’ve served on the HT program committee. Do you have any advice in this regard?
I’ve never been on the HT program committee, and I’ve never submitted a paper, so I’m working entirely on second-hand information. But I’ve been told that interesting papers about language-focused types of interactive digital media have been rejected not because they weren’t good papers, but because they weren’t “hypertext papers.” It’s a persistent enough rumor and perception around the field that if it’s not true (or would not happen these days) I think some effort should go into disabusing people of the notion (say, in next year’s CFP).
Finally, the question of textual focus is a good one to revisit. As mentioned above, I should perhaps change that to something about operating via textual logic. But Nelson’s early hypertext examples (chunk style and stretchtext) are certainly textual. And things that operate cinematically he calls, instead, hyperfilms. And he uses hypermedia as the umbrella term to include them all. So maybe hypertext really does have something textual about it. In fact, I remember when I was at the Hypertext conference in 1997 Nelson said something to the SIG planning group along the lines of, “Given the range of what you’re interested in, shouldn’t you be calling this the Hypermedia conference?” I think I’d tend to agree.
May 25th, 2004 at 12:55 pm
On references: yes, the hypertext conference tends to be a stickler for scholarship. I think the length constraints of the short paper form are also understood. You might find space for one or even two additional references by going back and rigorously pruning every word you don’t absolutely require. And, as you have done with Engelbart, some quasi-citations can be accomplished without a formal reference.
One reason the conference wants to see previous work handled well is the pathology you mention — the tendency in some circles to selectively forget prior work. If, as you say, “the most prevalent definitions in the literary community” limit “hypertext” to the node-link model to the exclusion of spatial, sculptural, taxonomic, and adaptive hypertext and hypermedia, then this error does need to be addressed.
On papers that “weren’t hypertext papers”: I’ve been a member of every HTxx program committee since 1989. Keep in mind that the HT conferences are extremely selective, accepting perhaps 1/4 of the papers submitted, and that they are so notoriously selective that, for the most part, chiefly good papers are submitted in the first place. In this context, and in an active research field, colleagues will occasionally disagree on whether a paper would best be presented at HTxx or elsewhere.
For what it’s worth, humanities papers have almost never been rejected for this reason. In the case of a few papers that involved substantial engineering work of general interest, in the context of a hypertext application that was felt to be less compelling than the engineering, it may sometimes have seemed best to encourage that the paper be sent to a different conference, such as SIGIR, CHI, SIGCSE, WWWxx, or Multimedia.
If this year’s CFP doesn’t calm whatever concerns people may have felt, I wonder whether those concerns can be assuaged.
With respect to the terms Hypertext and Hypermedia: I have always used these interchangeably, except that “media” as a collective plural sometimes introduces grammatical perplexities. Though the distinction among hypertext and hypermedia may have seemed useful in the era of ASR33 teletypes and 029 keypunches, by 1987 it was already clear that hypertexts would include all sorts of media. One of the hypertexts demonstrated at HT87, Amanda Goodenough’s INIGO GETS OUT, was completely visual, for example, containing no words.
The name might have been changed, as a matter of fact, but in the ’90s a different ACM conference series on Multimedia emerged, and the broader name would have created confusion and ill will.
May 25th, 2004 at 6:41 pm
Noah, the distinction between historical and analytical categories is pretty standard. You are a historical person, but that doesn’t mean that you’re not still alive. Analytical terms or categories, unlike historical ones, have to be formally definable. “Digital narrative” and “new media” are not analytical categories, but highly unfocused, ideological terms. You can do business with them, but if you try to do rigorous analysis based on them you quickly run into trouble. Just try to define them and you’ll see what I mean.
“Hypertext” is analytical or (merely) historical depending on how we choose to define it, or if at all. (But even if narrowly defined, it is still debateable whether it has to be computerized or not.) If hypertext researchers prefer to leave hypertext undefined in order to keep the borders of their field open, that is fine, but then the term becomes purely ideological, of the I-know-it-when-I-see-it kind. Kind of like “art”.
I would not be surprised if you discover that the hypertext community is not really interested in definitions or critical examinations of the term “hypertext.” It might create unnecessary tensions, and probably not result in consensus. Mark mentions “spatial, sculptural, taxonomic, and adaptive” hypertext — how is it possible to fit all those into a single definition? “Cool digital text stuff,” perhaps?
May 26th, 2004 at 10:09 am
Although I’m late to mention this, the distinction that Noah seems to be trying to make with his Nelsonian definition of “hypertext” is that between types of arranged media that are navigable in some way and more general computer programs. Hence the question of whether or not something has been “programmed.” It sounds like hypertext is that which sits between traditionally navigable documents (screens of text you can advance through, text that can be scrolled up an down) and creative programs that involve their authors in programming as much as they do writing.
Is it your sense, Noah, that Nelson wanted to make this distinction? If so, in reply to Jill’s question to me, I would say interactive fiction authored in Adrift, which eschews programming for something like the assembling of a StorySpace document or Web site, is hypertext, while IF created in general-purpose programming languages (Inform, TADS) is not. Of course, one can write simple programs or do complex arranging, so there is a difference between the system’s expressive capability and the activity of author/programmers within it. I am not sure where exactly one would draw the line, or how to make the distinction between programming and structuring/arranging/composing, but it seems like the beginning of a nice line of thought, albeit one that might make “programming” sound even more unnecessarily intimidating.
If the definition of this term that Noah undertakes is a “highly unfocused, ideological” one – and it may be, unlike the highly focused ideological ones that I and no doubt some others here prefer – it at least seems to highlight a dimension or aspect of creative work on the computer, one that is interesting. I think it could be the beginning of a definition that serves an analytical purpose, albeit one that is used for the analysis of authoring systems and development environments rather than finished works.
May 26th, 2004 at 10:56 am
I think this conversation has been a great help. It’s helped me figure out what my paper is really about (for me) and I think I’m going to write it again from scratch.
The real question isn’t, for me, how we find a definition (however “formal”) that will work for everyone. It is where we begin — when confronted with the “What is hypertext?” question — given that our most respected scholars give answers ranging from “structured knowledge work” to “merely a direct connection from one position in a text to another.”
That is to say, I’m still going to argue for attention to Nelson, but I’m going to start with explicit reference to the fact that the term has been taken in many different directions. Part of what makes attention to Nelson’s work useful is that it gives us a way of looking at these different ways the term is employed, and of understanding them relative one another, as movements in different directions from Nelson’s starting point. Knowing the starting point can also help us think about where we might want to move next.
To more specifically address Nick’s comment, I think the idea you propose is interesting, but it’s not how I read Nelson. Again, he doesn’t return to this topic (the way he does to “branch and perform”) so I guess we’re all speculating. But I think he was reacting to people saying, “So you’re gonna program some media into that computer?” And his response is, “I’m going to design, author, and edit some media, using the computer.” Programming might be part of the process (in fact, it’s basically inevitable) but Nelson thinks the same design/authoring language should be used as we would use for non-computer tasks of design/authoring. He wants to reserve the term “programming” for the construction of the system used for presentation (e.g., the operating system). While these systems are also, arguably, designed, they aren’t media. To put it another way, I think Nelson is saying, “Whether we use a pen or code to author media, let’s call that process authoring. If we call it programming we’ll put the emphasis in the wrong place.”
May 26th, 2004 at 3:06 pm
Whether seeking a historical or an analytic definition of hypertext, I think it makes sense that the definition include things that have always been called hypertext.
Two early systems seem pertinent to recent discussion. Rosemary Simpson’s GATEWAY (LMI, mid 1980s) had general computational power on the links but otherwise looked and felt like what would now be considered a familiar hypertext system. Since Simpson’s platform was the Lisp Machine, the generality of her links makes sense.
An even earlier system, Shneiderman and Borning’s TIES (and subsequently HyperTIES), did quite a bit of computation in the nodes. For example, it could automatically adapt references to units of measurment (pounds, shillings, and ounces) to the reader’s preferred format.
Espen says that, “Mark mentions “spatial, sculptural, taxonomic, and adaptive” hypertext — how is it possible to fit all those into a single definition?” I don’t see any need to go to the extreme of “cool digital stuff!”
Look at it this way: in 1989, we already had Storyspace, HyperSet, Trellis, and HyperTies — which we’d now call spatial, taxonomic, sculptural, and adaptive, respectively. But they’re all pretty obviously hypertext systems, and all have been regarded as such ever since. Nobody would mistake them for programming languages, or games, or operating systems, or what you will.
May 26th, 2004 at 3:39 pm
Fair enough, Mark. And the definition is..?
May 26th, 2004 at 4:33 pm
Espen, I’m just curious, but do you know if terms like “communism” and “psychoanalysis” have agreed-upon, rigorous, formal definitions? Would you say, if they don’t, that they aren’t analytically useful? Would you say (again, if such definitions don’t exist) that the best thing for the communities around them to do with their scholarly energy is to try to come up with such definitions, as historical understanding of the terms is insufficient as a basis for discussion?
Truth be told, I haven’t done much investigation of this sort of scholarly approach. I take it this general view is why I hear so many people trying to provide formal definitions for what is and isn’t a game? (It’s been puzzling me why that topic is so popular.)
May 26th, 2004 at 6:08 pm
All I am saying is that it must be slightly embarrassing for a scholar not to be able to provide a basic definition of the most central term in the field.
I have no idea exactly how communism or psychoanalysis are defined these days (nor do I have time to find out), but I don’t see why “historical understanding of the terms is insufficient as a basis for discussion.” We don’t need to define hypertext rigorously in order to discuss it critically as a historical, ideological term, or to see how it is used by the community.
May 27th, 2004 at 1:05 am
Ah, Espen, I think I better understand what you’re getting at. Of course I agree with you. After all, working toward a more historically-informed definition of hypertext is what my paper’s all about. I’ll post more on this thread once I’ve got a new draft (deadline looms).
May 27th, 2004 at 7:30 am
Just a brief comment this time: Of Nelson’s two definitions of hypertext, one (“text that branches,” let’s call it N1) has been very influential, and the other, while of interest to historians, has not gained any followers. It seems to me that this is for good reason, as Marie-Laure pointed out earlier. So while I find your endeavors to uncover the history of the term hypertext very valuable, I am less optimistic about your attempt to replace a successful definition with an updated version of a vague, forgotten one. If this is what the hypertext community really needs, I think they would have done it a long time ago.
In fact, there is a strange paradox here: Usually, the public meaning of a term, say energy, or even game, is less sharp than the scientific one. But, remarkably, with hypertext it seems to be the opposite. People outside or bordering on the field seem to have little problem defining hypertext (using some variation of N1), whereas inside the field we seem to find no strict conceptual definition at all.
May 27th, 2004 at 10:38 am
Espen, I’m sorry if I wasn’t clear, but I don’t consider “text that branches” to be one of Nelson’s definitions. Instead that’s lopping off part of one of his definitions and acting like it’s the whole thing.
As for the other issue, I address this in the original draft of the paper. It’s like Stalin’s definition of communism overwhelming that from Marx. In fact, in reading a bit of Marx, my impression is more of a powerful critique of capitalism than it is of a clear outlining of what would replace it (communism). Stalin’s vision of communism was clearer, and better known around the world, and so on.
Despite this, there are no scholars I have any respect for (in economics, or other fields) who operate from a definition of communism that starts with Stalin. For some reason, they all go back to Marx, who coined the term, and look at his somewhat-unclear definition. I could be wrong, but my hope is that our field will develop a similar care with its own intellectual history, even if it then is (for some period, or permanently) operating at odds with popularly-understood definitions of key terms.
Okay, now I really am taking a blog break until the new draft is done.
May 27th, 2004 at 11:00 am
Well, I would take a blog break, but now I’m struck by the fact that I’m being a bit slipshod myself in the comment above (as I was with the phrase “20th century thinker” in the first draft of the paper). Obviously, if I were being more careful with the term “communism” I’d also talk about Engels, arguably Cabet, and the fact that the term existed in the language (rather than being coined, as “hypertext” was).
May 27th, 2004 at 11:40 am
“non-sequential writing— text that branches and allows choices to the reader, best read at an interactive screen” –Literary Machines p. 0/2.
It definitely is Nelson’s, and the gist of it, is, clearly, “text that branches”. The rest is redundant.
May 27th, 2004 at 9:23 pm
Okay, I have a new draft of the paper. I’m not sure if it’s better or worse, because the new conception of the paper (arrived at through the helpful discussion here) is a bit more involved to explain. So while I think the new draft is closer to what I mean, I suspect that it suffers from its length limitation even more. I’ll mull it a bit and try to post some version soon.
Meanwhile, Espen, you and I seem the only people left in this conversation. Perhaps we’ve worn everyone else down? Has it come to seem too much about the two of us? Anyway, here’s a quick couple thoughts on your comment above and its quotation from Literary Machines (1981 edition? 1993?).
First, I don’t think that snippet of Nelson does much to support your 1994 hypertext definition (do you have a different one now?). There are many things that “non-sequential writing” could mean besides “a direct connection from one position in a text to another.” I don’t think the rest of the sentence (or of Literary Machines) is as superfluous as you say. “Branches and allows choices” sounds like “branches and performs” — chunk-style hypertext uses links to “branch,” stretchtext “performs” or “allows choices” in a way other than via links. Of course, LM is mostly about the Xanadu system, which was a publishing system envisioned to function via links and transclusion, so that should be borne in mind. (Do you think I should add transclusion to the definition as another way that hypertext can function besides “direct connection from one position in a text to another”?)
Second, even if I agreed with your interpretation of that snippet, I have the impression that people like Freud had differing explanations of the meanings of key terms over the course of their careers. For some reason, people still think it’s important to our intellectual history to pay attention to where the terms started. In fact, it would strike me as odd to argue otherwise.
May 28th, 2004 at 4:00 pm
For anyone who is interested, here’s the most recent draft. It may get a bit more revision before the end of the day, but probably not. I’m unhappy that I had to cut down the “implications” section — I agree with Jill about it being the most interesting part. But I’m glad that I was able to revise what lead up to it in the light of the conversation we’ve had here. My thanks to all.
WHAT HYPERTEXT IS
Noah Wardrip-Fruin
ABSTRACT
Over the past couple decades, as the term “hypertext” has gained a certain popular currency, a question has been raised repeatedly: “What is hypertext?” Our most respected scholars offer a range of different, at times incompatible, answers. This paper argues that our best response to this situation is to adopt the approach taken with other terms that are central to intellectual communities (such as “natural selection,” “communism,” and “psychoanalysis”), a historical approach. In the case of “hypertext” the term began with Theodor Holm (“Ted”) Nelson, and in this paper two of his early publications of “hypertext” are used to determine its initial meaning: the 1965 “A File Structure for the Complex, the Changing, and the Indeterminate” and the 1970 “No More Teachers’ Dirty Looks.” It is concluded that hypertext began as a term for forms of hypermedia (human-authored media that “branch or perform on request”) that operate textually. This runs counter to definitions of hypertext in the literary community that focus solely on the link. It also runs counter to definitions in the research community that privilege tools for knowledge work over media. An inclusive future is envisioned.
INTRODUCTION
How can we answer the question, “What is hypertext?” Our most respected scholars offer answers that span a wide range. In the literary community, the definitions offered often focus on the link. Marie-Laure Ryan [5], for example, states that, “In hypertext… the reader determines the unfolding of the text by clicking on certain areas, the so-called hyperlinks, that bring to the screen other segments of text.” Espen Aarseth [1] offers a similar view of the term, writing that, “Hypertext, for all its packaging and theories, is an amazingly simple concept. It is merely a direct connection from one position in a text to another.” In the computer science community definitions can sound somewhat different. In a forthcoming article a number of well known hypertext researchers from the University of Southampton [6] argue that link-based understandings miss “some of the more profound aspects of hypertext” especially “hypertext-as-interaction with information to build associations, and through associations to build knowledge.” In a similar vein, Peter J. Nürnberg’s [4] closing keynote for the Hypertext 03 conference (“What is Hypertext?”) offered “structured knowledge work” as a summary of the focus of the hypertext research community.
Given these incompatible definitions, we could argue that one group or another has more right to define the term. But intellectual communities must be home to arguments over the definitions of key terms — as these are arguments about the meaning of the field, necessary if we are to avoid stagnation. Better, instead, to understand the history of our terms, so that we may see how competing definitions of the moment are movements in different directions from a common starting point. That is to say, when we ask the question, “What is hypertext?” we are asking a question fundamentally similar to questions such as “What is psychoanalysis?” “What is natural selection?” and “What is communism?” We are asking about the meaning of an intellectual term coined by a particular thinker, around which further thinking has grown. While such terms are often taken in a variety of different directions after they are coined (much as Stalin took usage of the term “communism” in a direction that Marx is unlikely to have imagined) it is generally agreed that serious discussion of the meanings of such terms must begin with the work of thinker who coined them.
In the case of “hypertext” the term was coined by Theodor Holm (“Ted”) Nelson. This paper examines two of Nelson’s early publications of the term in order to provide a starting point for historically-informed definitions of the term.
HYPERTEXT, FILM, AND MEDIA
While Nelson may have presented the terms earlier, his first significant publications of the terms “hypertext,” “hyperfilm,” and “hypermedia” occur simultaneously — in the 1965 paper “A File Structure for the Complex, the Changing, and the Indeterminate” [2]. As the title suggests, this paper was primarily concerned with outlining a file structure (the Evolutionary List File, or “ELF”) inspired by Vannevar Bush and intended for personal use by knowledge workers. However, a final section makes a departure, noting that file structures like the ELF “make possible the creation of complex and significant new media, the hypertext and hyperfilm.” Two paragraphs later, Nelson expands the term “hypertext” in this famous sentence: “Let me introduce the word ‘hypertext’ to mean a body of written or pictorial material interconnected in such a complex way that it could not conveniently be presented or represented on paper.” In the following paragraph “hyperfilm” is briefly mentioned again: “The hyperfilm — a browsable or vari-sequenced movie — is only one of the possible hypermedia that require our attention.”
It is worthwhile to note the following: (1) “hypertext” and “hyperfilm” are coined within the same sentence; (2) both hypertext and hyperfilm are characterized as “new media”; (3) the larger category in which at least the hyperfilm is included is “hypermedia”; (4) while hypertext includes written and pictorial material, material that functions cinematically has its own term (hyperfilm); and (5) while what Nelson offers in this brief section does not explicitly contradict definitions of hypertext that focus on the link, links are not mentioned.
FORMS OF ‘HYPER-MEDIA’
Just as Nelson’s 1965 paper was initially concerned with the presentation of a file structure (rather than an explication of “hypertext”), his 1970 “No More Teachers’ Dirty Looks” [3] begins with a critique of concepts of “computer-assisted instruction” connected to his critiques of the dreams of artificial intelligence. Nelson’s proposed alternative to such systems is “responding resources.” He writes, “Responding resources are of two types: facilities and hyper-media.” On-screen calculators and graph plotters are given as examples of facilities. Nelson then writes of hyper-media:
Hyper-media are branching or performing presentations which respond to user actions, systems of prearranged words and pictures (for example) which may be explored freely or queried in stylized ways. They will not be “programmed,” but rather designed, written, drawn and edited, by authors, artists, designers and editors… Like ordinary prose and pictures, they will be media; and because they are in some sense “multi-dimensional,” we may call them hyper-media, following mathematical use of the term “hyper-.”
Nelson then presents examples of types of hyper-media that could be made available to students. The first of these is under the heading “Discrete Hypertexts.” Nelson writes: “‘Hypertext’ means forms of writing which branch or perform on request; they are best presented on computer display screens… Discrete, or chunk style, hypertexts consist of separate pieces of text connected by links.” This is the first appearance of the term “link” in the essay.
Nelson next presents a type of hypermedia called the “hypergram” (“a performing or branching picture”) followed by another form of hypertext — “stretchtext.” Nelson writes: “This form of hypertext is easy to use without getting lost… There are a screen and two throttles. The first throttle moves the text forward and backward, up and down on the screen. The second throttle causes changes in the writing itself: throttling toward you causes the text to become longer by minute degrees.” Note that Nelson referred to hypertext as “forms of writing which branch or perform on request.” Discrete hypertext uses links to branch on request. Stretchtext uses no links — instead making a non-branching performance.
Nelson’s essay continues, outlining further types of hypermedia such as the “hypermap” (a smooth zooming interface) and “queriable illustrations” (a type of hypergram). The fact that both “discrete hypertext” and “stretchtext” are situated within this list of examples of types of hypermedia leaves little doubt that hypertext is a subcategory of hypermedia. While the behaviors of discrete hypertext and stretchtext are quite different, what unites them is that they are “forms of writing.” While the examples shown of hypergrams, hypermaps, and queriable illustrations all include text, discrete hypertext and stretchtext are textual, are writing, in their mode of operation (even if they include pictorial material). Hypermedia, also, are differentiated from “facilities.” They are not tools, but media — “designed, written, drawn and edited.” From this we can conclude that tools such as spreadsheets and word processors are, along with calculators and graph plotters, not hypertext.
We can now, based on our examination of Nelson’s texts, provide the first two sentences of a historically-based definition of hypertext appropriate for a world familiar with the Web: “Hypertext is a term coined by Ted Nelson for forms of hypermedia (human-authored media that branch or perform on request) that operate textually. Examples include the link-based ‘discrete hypertext’ (of which the Web is one example) and the level-of-detail-based ‘stretchtext.'”
IMPLICATIONS
How should a historically-based definition of hypertext continue, after these initial sentences? It would make sense to note that, in the literary community, the definition of hypertext shifted so that it applied almost exclusively to chunk-style media. It could be speculated that this took place because most authors who called their work hypertext fiction or poetry worked in link-oriented forms (though exceptions such as the work of Jim Rosenberg were well known). Within the hypertext research community a different shift took place, with a focus on knowledge work (largely on tools Nelson might have instead called “facilities”) rather than media. At the same time, this community maintained a definition of hypertext’s possible structures that was broader than the chunk-style. With the rise of the Web (a chunk-style hypertext media system) the term’s popular understanding shifted.
Going forward, those of us who discuss hypertext, in any community, must decide whether we want to acknowledge this history. If we do, those working in the literary community must reconsider hypertext definitions focused on the link, and those working in hypertext research must reconsider definitions that privilege knowledge work over media. We may find, in such historically-motivated broadening, our future.
REFERENCES
[1] Aarseth, E. Nonlinearity and Literary Theory. Hyper/Text/ Theory, 51-86. Baltimore: Johns Hopkins Univ. Pr., 1994.
[2] Nelson, T. A File Structure for the Complex, the Changing, and the Indeterminate. Association for Computing Machinery: Proc. 20th National Conference 1965, 84-100.
[3] Nelson, T. No More Teachers’ Dirty Looks. Computer Decisions 9, 8 (Sep. 1970) 16-23.
[4] Nürnberg, P. What is hypertext? Proc. Hypertext 2003, 220-221.
[5] Ryan, M. Narrative as Virtual Reality. Johns Hopkins Univ. Pr., 2001.
[6] schraefel, m. c., Carr, L., De Roure, D. and Hall, W. You’ve Got Hypertext. J. of Digital Information. Forthcoming.
May 29th, 2004 at 8:07 am
Dear Noah,
Apologies for the late posting — I actually had to give my first lecture on Hypertext!! ( And so I found this discussion quite helpful.)
Although this post is post-presentation I felt it necessary to clarify my query. Noah, when I asked about the ‘bounds’ of the ‘linking’ I was referring to whether texts that ‘link’ in some way outside of the present text would qualify as hypertext in the ‘discrete’ sense? There is a difference between electronic texts that have links to pages that are within a storyworld and the current website and those that are links to pages outside of the current website but still in the storyworld. For instance: a hypertext fiction that the user reads and must click on words that open up pages that continue a sentence of plot within the website as opposed to a fictional website that has links to another fictional website.
It is clear to me now that what you were addressing in Ted’s words was the nature of the text in current view: that is, a text that has links. It matters not, therefore, where the links go to. You were focusing on a text having the characteristic of having links rather than “separate pieces of text connected by links”. It is a difference of perspective.
My query was based on my own research interest which is in alternate reality gaming, transfiction and multichannel storytelling. I am working on a poetics (oh that word!) of multichannel storytelling (which is in fact a navigational design). That is why I was interested in whether you saw Ted’s definition of ‘hypertext’, specifically ‘discrete’, as being a description in some sense of the explicit connection between media and texts. I think it does have application despite its current usage.
I hope you get to read this and I hope I have explained my query further so that you may respond.
June 7th, 2004 at 1:31 am
Christy, now it’s my turn to apologize for posting late.
I’m going to try to paraphrase your question. Let me know if I’ve got it right.
I think you’re asking whether I think it matters, from the point of view of Nelson’s concept of “discrete” or “chunk style” hypertext, whether the pieces (the chunks) are all made available through one website or are distributed across multiple websites.
It sounds like you’re asking this, specifically, in the context of works that foreground this distribution in the reader/player experience.
I think you’re right that there’s a difference in this that’s worth noting. But I don’t think the difference is between being discrete hypertext and not. I think such work is still hypertext, and of the node-link variety, whether the contents are distributed across multiple sites or not. The fact that the distribution is emphasized for readers/players seems like it’s operating at a different level.
Is that how it seems to you, as well?
June 7th, 2004 at 1:35 pm
There’s no way to permalink, but Diane Greco’s June 2nd entry has some interesting thoughts related to this thread. (And, if you want to know why there are no permalinks, scroll down to May 31st.)
June 11th, 2004 at 1:13 am
Yes, there is a difference that is worth noting, especially since distributed storyworld content is becoming an active genre in itself. And yes, the term does apply – I just wanted to make sure it was an acknowledged usage.
I particularly wanted to check the usage to see if it is ‘ok’ to place the multichannel navigation/spatial model under the area of hypertext, since I’m finding the model can be applied in navigation within a channel as well. This ‘local’ space can be within the channel (distributed content within a channel) and within works that are hypertextual (intra local) — with ‘neighbourhood’ being movement to another channel. Yes the difference is with the reader/user but also the writer/designer – which places the concerns within the creation and usage level(s).
Cheers,
christy
May 25th, 2004 at 2:01 pm
What is really of interest, to this reader, is the emphasis on hypertext as a “responding resource” … but until the whole process can be made as simple as basic writing, or drawing, or speaking, then the dynamic qualities of our natural…
June 30th, 2004 at 2:27 pm
The paper was accepted, I’m happy to say. Thanks again to everyone for their very helpful feedback — especially Espen, who stuck with me through the rough patches.
Cristy, sorry to be so slow responding. I’ve been on the road quite a bit. Hopefully I’ll have a considered response in the next few days.
July 2nd, 2004 at 8:18 am
Heh; don’t forget to dare the conference to define their key term. It is hytime they did.
July 3rd, 2004 at 1:52 am
Espen, good idea. Maybe I should point them toward DiGRA’s recent consensus on the meaning of “game” as a model for the path they might take while working it out? ;-)
July 3rd, 2004 at 6:39 am
Well, since games are (or used to be) a form of hypertext (according to previous HT conference CFPs) I am not sure such a suggestion would strike the right chord.
But then again, the vigor with which game definitions are tested and contested really energizes the field. Consensus is overrated; the path is the goal.
August 4th, 2004 at 11:28 pm
At 10:45, in a session called “Foundations,” I’m giving a paper that I imagine will raise a few eyebrows. It’s titled “What Hypertext Is” (final pdf) and it was revised extensively after a productive discussion over at GTxA.
October 10th, 2004 at 7:21 pm
Regarding communism. The word was originally associated with the Icarian Movement of tienne Cabet. Karl Marx and Friedrich Engels did not coin the term, but rather picked it up from Cabet and German Weitling.
February 15th, 2007 at 4:03 pm
[…] nd animation. Can the player use the mouse to point and click on items? No. That would be hypertext. Can the player talk to the machine instead of us […]