Category Archives: Nelson

How to speak to a computer

“everybody In computers has deeper motivations and interior twists that form his own special ties to these machines; and when It comes to our choices of fantasy machines, obviously an even deeper level of psychic imprint la projecting Itself Into the world.”

Latour, in his critique of scientific objectivity, argues that a large majority of science is simply problem solving.  Without a problem to be solved, there is no scientific work to be done.  Nelson’s computer dreams made me reflect on this idea based on the degree of whimsy he describes the computer engineers to be acting on. This is to ask what is the result if we look at computing the same way Sir Edmund Hilary looked at Everest?  Is “because its there”  a good enough reason to create and deploy these untested visions?


In this light, I see a strange world of computer science, one that is funded to build better bombs and police state control, but is staffed by human dreamers who wish to make their imagination a reality.

Nelson’s advice is to

“sort out the dreams and put them on hangers so you can try them on, and maybe choose an ensemble for yourself.”

I wonder if this self reflectivity is far enough or even explains the complexity of the tensions between the visions of these technologies and their realities.  Instead, when I observe the backlash (In Europe) over GMO’s and early fears over bioengineering I see a conflict between un-involved publics and those who hold the keys to the direction of future technologies.


As an aside, I see an interesting divergence of Nelson’s critique on the future of information retrieval.  Although it is true that Google search would fail to answer his question  “What did Hegel say about determinism”, folks wanting to know the answer would use different inputs they would feel would deliver better results like “Hegel” “Determinism” “early works” “counter argument”. In a way, we have learned to speak like a computer program to maximize the output.

Full Nelson

1. Nelson, like many of the alternative thinkers we have read about so far, spills a good deal of ink on education and the possibilities for education reform using technology. Why is education almost invariably chosen as a means of applying these authors’ ideas? What is it about the educational endeavour specifically that makes it so conducive to both criticism and the application of technological ideas?

2. What potential downsides can you see to two-directional linking as envisioned by Nelson? How might the WWW have evolved differently if it had been implemented in this way?

Cultural Evolution

I have noticed two themes in our semester’s readings. One is about augmenting ourselves with technology. The other is moving ourselves away from excessive technology. While one is a push for advancing the culture and efficiency of the human race, the other is a receding to the pure essence of what it means to be “human” – the “primary wants” and “individualization” of humans (Gandhi, Schumacher, Tagore).

Ted Nelson falls into the second category. His solution, hypermedia, augments and makes more efficient the process of organization, ideation, writing, and even education. Such hypermedia transcend the linear “experience” of which physical paper documents are characterized.

Nelson distinguishes file structures used for creativity and those used for data processing; the former being that which he designs “hypermedia”.

Contradiction: If Nelson’s idea heightens efficiency in creativity and problem solving, then why do so many creative writers prefer using traditional paper and pen? Why is Nelson’s complex organization tool not in pervasive use for creating fictional masterpieces?

Perhaps the way a traditional book is read and the means in which a writer writes are successful because of what the second category of our readings touch upon: they are more “human”. More simply, perhaps our minds have been selected, evolutionarily speaking, to think linearly and to write linearly. After all, men have, even before the rise of writing, told stories in a linear, narrative fashion. This harkens back to some of our first civilized societies. Our sentences are crafted linearly. We communicate with one another in a linear fashion; one thing comes after another.

The core of Nelson’s design is that which is “hyper” – complex and interrelated. And perhaps this works with some domains of work, recollection, and retrieval. But for domains of creative writing and personal reading, the linear, the “un-hyper” dimensions, seem to reign supreme. It just seems more “natural”, more “human”.

The act of writing on paper, which is so personal, lends to creativity and memory making in a way that technology does not come near.

A parallel exists for Nelson’s perspective on education. In a diagram in Dream Machines, the teacher and computer is seen as a barrier to a student’s understanding of the subject matter. The solution? Hypermedia (which he emphasizes in all caps). Education and proper learning is augmented by technology, not a far cry from his solution to solving our poor organization and writing habits.

This question has been running through my mind the past couple of weeks:

There seems to be a conflict between two different categories of thought:

1) Being more “human” (Gandhi’s “Primary Wants of Man” and “Individuality of Man”, Schumacher’s “Buddhist Economics”, Tagore’s “Creative Freedom”)

and 2) Inventive and futuristic technology to augment human capabilities and encourage thinking “Outside the Box” (like Nelson’s hypermedia)

How, as designers, can we leverage these two categories? Should we design technologies that elegantly accompany the evolutionary biologies of our brain and behavior (What it means to be human), or should we design technologies that augment and expand the “human” experience further than that of “social animals”?

“The last thing to be ruined determines your profession.”

Nelson’s piece on education in Dream Machines is pretty interesting.  While I like the idea of letting students dictate their path, decide when they’re prepared for tests, and generally take more autonomy in the classroom, I’m not sure its realistic. Nelson writes,

“Motivate the user and let him loose in a wonderful place.

Let the student control the sequence, put him in control of interesting and clear material, and make him feel good — comfortable, interested, and autonomous. Teach him to orient himself… Such ultra-rich environments allow the student to choose what he will study, when he will study it and how he will study it, and to what criteria of accomplishment he will aim. “

That kind of transformation would require both a fundamental change in teaching and an equivalent — or larger — change in students. Education here is built upon teacher-as-head, slow, steady, ‘all move as one’ didacticism. Maybe younger students are pliable and would be able to cope with a change to an more enlightened teaching — but the thought of the chaos of that initial transition is a little scary.

Nelson’s “fundamental point”:

“Computer assisted instruction, applied thoughtlessly and imitatively , threatens to extend the worst features of education as it is now.”

is reasonable. This seems to be a trap some large webcast courses and MOOCs fall into. Are there examples of those that don’t? What are examples of good integration of technology (as defined by Nelson) in public schools?

What’s wrong with stories?

Many of Nelson’s concepts for future information retrieval allow us to non-linearly and non-predictably traverse text. In “Some Premises Relevant to Teaching,” he says claims “there is no natural order or sequence” of teaching. Hypertext, hypermaps, hypergrams and his other IR suggestions give the user a tremendous amount of choice about where they move.

Where do stories, as a tool for teaching, fit into this view? Isn’t it sometimes valuable that to force ourselves to listen to facts, presented in a certain order? Is it possible to have stories that don’t imply a “natural order or sequence”?

I’m skeptical that our brains can handle as much “on the fly” connecting and arranging as Nelson claims they can. I personally find clear sequences of facts and texts very helpful in learning a subject. I also know that there are fundamental limits on our short-term working memory that make thinking *harder* when faced with complex information, like the ones Nelson describes.


Thirteen years he dedicated to these heterogeneous tasks, but the hand of a stranger murdered him–and his novel was incoherent and no one found the labyrinth.

the dreamfile: the file system that would have every feature…

in many ways, Nelson’s writing reminds me of the 19th and 20th century utopian thinkers who would carefully design (on paper) their brave new cities, occasionally making forays into building of them. much like Nelson’s experiments with links and filestructures, none of these utopias were lastingly used.

computers, once a branch of mathematics, are now their own field (but the development of fluid logic indicates a possible merger with the art of wind instruments).

like with the pile of discarded utopias, it’s worth wondering why Nelson’s ideas never found lasting traction. his idea of two-way links had very broad critical appreciation (and it’s still championed by thinkers like Jaron Lanier, who also happens to love wind instruments). yet the system we all use is Berners-Lee’s haphazardly-constructed HTTP, done as a contract job during his tenure at CERN.

yes, ARPA took up the ENQUIRE project and promoted it, yadda yadda, but why did they take up that system?

The author of an atrocious undertaking ought to imagine that he has already accomplished it, ought to impose upon himself a future as irrevocable as the past

perhaps writing about plans is the whole problem. to my knowledge, Berners-Lee never published anything about hypertext systems. he simply built one. perhaps the HTTP/HTML-industrial complex gained traction simply because it existed.

why do you think Nelson’s ideas about two-way links didn’t catch on? why do you think they haven’t been implemented by someone else?

A Garden of Forking Paths & You- What Technology Is, Is Not, and Might (Never) Be

Ed: I kind of just kept writing for longer than I expected.  Apologies for the long post! 

The tech community has taken and run with more than one Borges short story over the years (for another example of an influential story, see The Library of Babel).  And I can’t blame them: Borges writes some pretty cool stuff.  But in a world where the Web is ubiquitous, it’s easy to overlook the the sweeping influence that The Garden of Forking Paths and the system of hypertext that it inspired has had on the world.  By the same token, it’s easy to miss the distinctions between what hypertext is, what it was intended to be, and what it might (not) become.

The Garden of Forking Paths, itself, is far from a technical specification for hypertext.  It is a meditation on possible futures, forking timelines, agency and an unfinished novel of infinite possibilities.  But it is precisely that novel, branching into all possible futures rather than flowing toward one or another, that is often credited as the inspiration for hypertext.  Indeed, Nelson’s proposed system, underneath the technical description, pretty closely mirrors the book in Borges’s story.  An arbitrarily large collection of ordered lists of text (paths), arbitrarily linking to one another (branching), allowing for lists to link or converge, and for a user to trace a relatively ordered path through a series of interconnected pieces of text.  Nelson’s vision is optimistic – writers will be able to drastically cut their workload, annotations built upon annotations will allow us to disambiguate our work to any desired level of precision, and changes made in one list will propagate effortlessly into anything related.

The reality of HTML and the world wide web is an echo of what Nelson envisioned.  Many of the pieces are there- The web is an enormously complex series of linked pages, the freedom to interconnect disparate pages laying the foundation for the unprecedented flow of information that the internet has allowed around the world.  Yet for all the convenience the web has allowed us, it is difficult to say that much has significantly shifted about the way we compose or consume information.  As far as I’m aware, (long-form) authors have move mostly from writing linear paper manuscripts to writing linear digital manuscripts.  Wikipedia, perhaps the poster child for a source of robust, interconnected information, feels more like a traditional encyclopedia with conveniently locate-able citations than anything thoroughly different than its ink and paper forebears.  Though I might be fooled by my dearth of pre-internet experience, I cannot help but feel like I’m waiting for the other shoe to drop.

Nelson’s Dream Machines, as tangled and confusing and immature as it can seem, provides a good counterpoint the the very same optimism which Nelson expresses in his earlier paper.  Much of the text is spent just describing what we do and don’t know about computing, and largely leaves the reader to draw conclusions.  But the overarching slant of the piece pushes back on the sort of technological futurism that we saw with Engelbart and Licklider.  Many ‘prophets’ of technology have forecasted drastic shifts in humanity and technology as time passes – from Licklider’s human-computer symbiosis to Kurzweil’s singularity.  Technology and its supposedly relentless advance have at times been charted as some sort of linearly or exponentially accelerating force of nature (See this ridiculous graph, which equates the development of Eukaryotic life and personal computers).  Similarly,  Nelson points to what remains a chorus among many computer scientists: “We can’t do that yet.”  Early in the reading, Nelson makes a simple and obvious observation: perhaps the potential of technology is not so limitless as it is so often portrayed.

In many of the readings we have done to date, industrialization and the rise of technology have been portrayed as an inevitable and fundamentally transformative force.  Reactions to this vary – Gandhi argues for a return to the Charkha, Tagore points toward a loss of fundamental wholeness, Licklider and Engelbart excitedly imagine unlimited possibilities for human evolution.  Always there is this implication of fundamental change – a basic shift from some (previous) model of human experience to some new and {improved/different/worse} model. Yet it is this inevitable change itself which I find questionable.  I accept that technology is a magnifier of certain parts of the human experience – a greedy man can exploit more people through the mechanisms of a modern corporation than he may have been able to as the head of a guild;  the ubiquity of the internet allows people to spread thoughts or ideas faster than ever before; the mass production of goods has vastly widened the potential proliferation of certain products.  But I fail to see what has been fundamentally altered.  We are born, we live, we die.  Somewhere in between we probably love, and laugh, and hurt, and do all of those other things that people tend to do.

It’s undeniable that the advent of technology has come concomitant with sweeping changes in the structure of our societies and our interactions with our environment.  But I fear that we are missing the tree for its leaves.  It isn’t the factories that are polluting the Ganges – it’s the people who built them.  As far as I can tell, technology hasn’t much changed us – it has simply changed the scale at which we operate.   I remain unconvinced that the a shift in the use of technology can redress the fundamental issues of human nature.  I don’t mean to imply that we should just give up and let what happens happen – there is something to be said for preventing the proliferation of nuclear weapons, or other dangerous technology.  Exerting human agency over the use of the tools we build is not just important, it is essential.  But it seems disingenuous to frame the ills of human nature in the tools we use to accomplish those ends.  At one point in Dream Machines, Nelson asserts that we might not want computers that talk back to us.  I’d ask: “Does it really matter?”

Perhaps I’m missing something important.  Over the course of the semester, we have read a variety of views on what implications technology holds for the human race.  But what does it take to really, actually, change the human experience?  What does that mean? Are we fundamentally different than we were in centuries past?  Are we counting down the days to the technological singularity, or are mass production, industrialization, and personal computing simply a few more incremental changes to the environment in which we continue to be human?