Wednesday, September 23, 2009

Crazy Man, Crazy Vision

“Legendary was Xanadu” -Newsreel Narrator, Citizen Kane

Much like Charles Foster Kane, Ted Nelson also had big aspirations for a monumental project. And like Kane, Nelson’s Xanadu was never completed. But in the nearly 50 years since Nelson’s project was born, it brought many concepts and ideas into the computer world that have shaped modern computing. Without his eccentric (for the time) visions and ideas, today’s computing landscape may very well look different than what we are used to.

In the beginning…

Ted Nelson must have been a character to behold. A rebel and nonconformist from a young age, Nelson suffered from (unknown at the time) Attention Deficit Disorder. In fifth grade, after backing out of an attempt to stab his teacher with a screwdriver, he coined the saying that would guide his life: “most people are fools, most authority is malignant, God does not exist, and everything is wrong” (Wolf, 1995). His rejection of the norm from grade school all the way to college, and always had his own way of doing things. For better or for worse, it was this spirit that led him to visualize wild concepts and relentlessly push for them where others may have not had the vision for such radical ideas.

The seeds that would later try to grow into the Xanadu project were planted in 1960 while Nelson was in graduate school at Harvard. For a term project during his freshman year, he attempted to create a word processor, though unlike many of the processors of the time, his had many unique features. His design would contain features that allow users to compare text side by side, version control, and revision by outline (Wolf, 1995). Setting a trend that would continue through his life, the project was never completed. The project quickly developed into his brainchild. The idea of hypertext was developed (though it would not be coined by Nelson until 1965), as was his plan of storing all information ‘online’ and accessing it through a computer. He recalled in his (unfinished) autobiography, “As those weeks of fall 1960 passed, I tried to tell everyone my growing vision. It was obvious, unified and cosmic. We would not be reading from paper any more. The computer screen, with its instant access, would make paper distribution of text absurd. Best of all, no longer would we be stuck with linear text, but we could create whole new gardens of interconnected text and graphics for the user to explore” (Nelson, 1993). He followed that by stating that “books would be obsolete by 1962”. With statements like these, it’s clear to see why everyone thought he was crazy.


With a reputation like he was starting to build, it would be a crime if any project he created did not embody all his ideals. In that regard, Xanadu did not disappoint. His mission was simple, yet more complex than he himself realized. The basic idea was to catalog information into a server, make hyperlinks between related information to allow for non-sequential reading, and allow users to seamlessly search access the information. The full list of the project’s ‘requirements’ or rules can be found on the FAQ site ( These rules are very vague and generalized, and don’t provide a great sense of what Nelson’s vision for Xanadu was. A better description can be found in one of the Chapter Threes (yes, there were several chapter threes, as well as several chapter ones, though only one chapter zero and two) of Nelson’s book, Literary Machines: “[The Xanadu Hypertext System] permits the publication and instantaneous world-wide delivery of interconnected works...will run on any personal computer…front ends of any kind are possible” (Nelson, Literary Machines, 1992). The chapter goes on to discuss many of the other capabilities of the system, which I do not have room to go into here.

On the whole, Xanadu is peculiarly similar to Vannevar Bush’s Memex, as proposed in his article “As We May Think” published in 1945. Indeed, Nelson was highly influenced by the article, so much in fact, that it is reprinted in its entirety in one of the Chapter Ones in Literary Machines. While, in 1945, Bush’s writing was mere fantasy, Nelson’s concept had (a few) more roots grounded in reality and what was possible. However, even the Xanadu vision was far ahead of its time, a chief reason that finding investors to fund the project were nowhere to be found. Every time that a pitch for the project was made, investors nodded their heads and were never heard from. There was also the issue of Nelson’s limited knowledge of programming and math. Not being a developer himself, he often set development goals that were too high for his small team and for the time. That isn’t to say he didn’t have everything planned out. On the contrary, the book Literary Machines is a complete description of the system and business plan for how the Xanadu project was to be run, down to what employees would wear and the marketing of the product.

The problem wasn’t the incomplete vision, but rather a series of development issues. For one, funding was an ongoing issue (until acquisition by Autodesk in 1987). . Funding the machines on which to develop was also a considerable expense that was at times unable to be met. The project was also never able to attract talent to stay for any length of time, so the small but talented, team was constantly changing. Changing nearly as quickly was the technology, both hardware and software, and combined with the slow pace of development meant that the team was switching computers and languages more often than they wanted. In short, Xanadu faced enough problems to kill nearly any other endeavor, but the strong belief of Ted Nelson in his vision kept it going setback after setback.

What it gave us

The description of Xanadu in the section above outlines some of the underlying concepts behind the way the internet, or at least certain parts of it, specifically user generated content such as blogs and wikis, function today. Hypertext is the backbone of the internet where it is no longer common to just post static information in web page ‘dead ends’. The linking from blogs to a dozen different sources per post allows the user to create their own dynamic, nonlinear experience. Nelson’s later coining of hypermedia, the linking in multimedia, has really been making great advances in recent years, with clickable video. This allows either clickable links to be laid over a commercial or people to pinpoint clickable links to points of interest in a YouTube video. To imagine the internet without these things wil soon be impossible; hyper linking completes the notion of what the Internet and modern computing mean to us.

The car, airplane and the telephone made the world a smaller place in the last century. Hyper linking has made the internet smaller and more manageable in the last decade. With the help of search technologies, we can follow thought paths through the internet, seamlessly and instantaneously jumping from place to place, from one idea to the next. Creating new content on the internet in blogs or on social networks is simple and instant; according to a 2008 survey, Facebook logged 161 million unique users in one month, for example (Schonfeld, 2008). That number can only be higher a year later, and that’s only one social network site; there are dozens of blogging sites and social networks out there. These users are posting links to stories they like, interesting videos, and writing about their lives. Some of the stories may be dead ends, but many lead to places where further hyperlink exploration can occur.

Ted Nelson’s vision for hyperlinks appears to be realized today. However, his idea had something that today’s internet decidedly lacks. Everyone that has used the internet for any amount of time has run into a dead link. The dreaded HTTP 404 is what Nelson’s concept was going to avoid through a unique system of having a master numbered catalog of all the documents. This issue is one of the major shortcomings in today’s system. Regardless, as the internet boom of the late 80’s and 90’s hit, the Xanadu team realized that their struggling project was quickly becoming irrelevant. The internet offered, while quite simplified, a version of the team’s universal library. Some of the more specialized features of Xanadu, such as version control and comparison are being implemented in more specialized internet and desktop apps, where they are more relevant. What is incredible is that Ted Nelson realized the potential of these features before the world had anything that resembled usable word processor.

* * *

Xanadu was a colossal project envisioned far before its time, and regrettably plagued with development issues for dozens of years, leading to its ultimate demise. It didn’t die without its influence being spread all over the computer industry though. The ideas of Ted Nelson gave us the concept of hypertext and hypermedia, leading to the interactive internet we have now. Without a doubt, Web 2.0 could not have happened without the tight, hyperlinked web that allows users to post, link, and re-link content. Users’ ability to freely share their content instantly worldwide was Nelson’s dream, and to an extent it was realized. Also realized were Nelson’s original ideas of text comparison and version control, both of which are the backbone of modern source control systems. The influence and power in his ideas are overwhelming, especially considering the time in which they were thought of. Nearly all who worked on the Xanadu project believed in it. But the lack of leadership and money made it impossible to realize. To think what would have happened if both those factors were present in the 60’s and 70’s is pure speculation, but I wonder if the internet, like what we know today, would not have come a decade or two earlier. In the same way, it is curious to contemplate what would have happened if Ted Nelson was a normal child and went through school like everyone else. Would his ideas still have come to fruition? Would they be different? How would our digital world look like? We’ll never know, but one thing is for sure: the impact he has made with the Xanadu project is being felt by all today and is driving the internet and digital knowledge forward.

Works Cited

Nelson, T. (1992). Literary Machines. Sausalito: Mindfull Press.

Nelson, T. (1993). unfinished autobigraphy.

Schonfeld, E. (2008, 10 29). How Many People Use Social Netowrks? Retrieved 9 22, 2009, from TheDigitalSanctuary:

Wolf, G. (1995). The Curse of Xanadu. Wired .

Yeggor Pavlov

No comments: