Jumping-Off Point

Today in Tedium: The language of the internet is ever-shifting. But that said, it’s not often I come across a term from the days of the early internet that I’m completely familiar with. Usually this happens with really new terminology, not the old stuff. If I find myself intrigued by a term like webring, wikitorial, or free-net, it’s generally because I remember it and think it would be something folks like you would like to read about. But when I heard the term “jumpstation” recently, I initially thought that it was a new way to say “trampoline.” In fact, it’s actually a reminder that the terminology that defined the early internet was not necessarily set in stone—until it eventually was. Today’s Tedium tries to explain what a jumpstation was, and what it says about the internet we left behind. — Ernie @ Tedium

Not a trampoline, but it works kind of like one. (Hans/Pixabay)

Why the jumpstation concept actually made sense during the earliest days of the world wide web

The jumpstation concept is fairly simple—effectively, it’s a collection of links on a topic, sort of a combination of a portal and a webring. In many ways, it’s like an FAQ, or a directory of links. Sort of the secret sauce of the jumpstation is that it directly pushes you in the general direction of the type of content you’re trying to read.

It’s basically a less effective search engine, the result of the fact that during the earliest parts of the World Wide Web, search interfaces were very much not a given. I remember some of my earliest web experiences, using a terminal screen and dialing into Lynx, where the internet was totally text-based—and because of limitations baked into the service, which I had no way to pay for because I was 13, I could not put in my own URL. I basically went as far as the public links could get me. There was no Google; hell, Yahoo had just been created at this point, not that I had any way to access it.

Early on, the experience of surfing the web was very much comparable to Gopher, in the way that general structure made up for the lack of an easy way to find anything. If anything, the jumpstation was the web’s way of recreating the Gopher experience.

Of course you needed books to use the internet, whether or not they were written by Michael Wolff.

It seems like a concept created for a time before we had figured out any of the basic tools that would make the Web worthwhile on a mass scale.

There are a lot of confusing elements of the jumpstation concept to me, in retrospect. References to it are generally sporadic, mostly found in books and newspapers during the late 1990s and early 2000s.

But perhaps the most confusing part? One of the earliest examples of the very thing that replaced the jumpstation, the search engine, was actually called JumpStation.

And nearly as confusing: Even after the search engine became common, the jumpstation terminology got applied to the very thing JumpStation was trying to kill.

Even if we don’t call things jumpstations anymore, it’s still interesting to consider why we once did.

(Agence Olloweb/Unsplash)

How JumpStation, the search engine, died before it even had a chance to take over the world

In the early years of the World Wide Web, technical students at universities around the world were trying to figure out ways organize and make sense of this tool that was borne from the mind of a researcher and given a leg up by a fellow college student.

In a lot of ways, it makes sense: In its earliest iterations, the internet lived on college campuses, where it was well-suited to be a hotbed of experimentation. A lot of examples of this experimentation turned into major parts of digital culture: Yahoo, the creation of two Stanford graduate students that started as a giant directory of web pages, was formed around this time.

But the problem that Yahoo solved, discoverability, did not at first include a search element, and that meant a whole lot of clicking before you found anything good. (Famously, Yahoo later added one.)

Fortunately, lots of other college students from around the globe were working on this problem, which many saw as imperative to the future growth of both the web and the internet in general. In a 1995 book titled The Mosaic Navigator: The Essential Guide to the Internet Interface, writer Paul Gilster laid out the case for why search engines were destined to usurp link directories:

Although we often find what we want, the process can take quite some time.

Let’s be clear on the reason why. Although the Web’s advocates speak of it as the ultimate information engine, it nonetheless possesses the same limitations as any network tool. Hyperlinks are, indeed, a wonderful way to connect to related information, but they must be built by hand. That means a human being must decide where the hyperlinks go and how to place them logically so that the reader can draw maximum utility out of the information space they create.

And even if we know where the good sites are, the entire system rests upon its connections to other information; the Web is meant to be a seamless whole. Choose a particular hyperlink and we may easily wind up on a home page half a continent away, one that has been built by someone whose sense of order and relationships differs substantially from our own. Then, too, following even the best of links to get to what we need often means starting with the general—archaeology as a subject, say—and working our way inexorably to the particular—the Incan ruins at Machu Picchu. It is this forced narrowing, requiring us to start with the broad and move to the narrow, that so frustrates the long-time user of the Web.

What we need is something that brings the focus of a search engine into the browsing environment. We need to be able to specify exactly what we need and let computers do our searching for us. They would then return a list of sites, World Wide Web pages that conform to our specifications. At that point, we could use Mosaic to access these pages and find, we hope, the promised data. The good news is that such search engines are beginning to appear on the Web’s horizon.

It’s with this in mind that a variety of tools were built using an automated tool called a “robot,” which basically scoured every page it could find on the internet, digging through link after link until it found something good.

One of the earliest examples of this came from Jonathon Fletcher, a graduate of the University of Stirling who found his PhD plans stymied by a cut in funding and ended up taking a job with his alma mater. At the time he graduated in 1993, the early rumblings of the World Wide Web and NCSA Mosaic were just creeping out into the broader internet, and he was in a role—in charge of building a web server for his university—that allowed him to see the bigger picture of what the internet needed at the time.

“If you wanted to see what had changed you had to go back and look,” Fletcher told the BBC in 2013. “With a degree in computing science and an idea that there had to be a better way, I decided to write something that would go and look for me.”

An early screenshot of JumpStation II, the second iteration of the search engine.

It was useful to far more people than just him. He was one of the first developers out of the gate with a web crawler, a robot that looked for new pages online, automating the discovery work that had once been done manually. Fletcher added the ability to search the results of the crawling to see what cropped up—effectively creating a search engine for the web in the process.

But JumpStation, one of a handful of experiments of its nature at the time, ran into issues of university politics, just like another university-borne internet project of the era, Gopher.

Fletcher had created a technology that would later change the world in different hands—that major research facilities would soon improve upon—but he couldn’t get the budget to continue supporting it. Despite its high popularity on the internet in 1993, he was effectively running a popular resource on shared hosting, meaning that it was too weak to even parse full websites—it could only analyze titles and headers. So, eventually, he left for a job in Tokyo, a decision he says made sense for him at the time, even if it does give him a tinge of regret.

“I was obviously not very successful in convincing them of its potential,” Fletcher told the BBC. “At the time I did what I thought was right, but there have been moments in the last 20 years where I’ve looked back.”

The result of this failure of vision was that the University of Stirling did not financially benefit from the creation that happened on their campus, like other schools with early internet-based technologies had—like the University of Illinois, which had fostered PLATO, Eudora, and Mosaic on its campus.

Within a few years, competitors like Lycos, AltaVista, and Google appeared on the scene, taking Fletcher’s basic idea—which he only received praise for years after the fact—and turning it into big business.

Ironically, despite creating a search engine intended to replace long lists of links, “jumpstation” became a term applied to those lists. Perhaps it’s because JumpStation, the search engine, lived such a short digital life.

The Federal Acquisition Jumpstation.

Strangely, an obscure part of the U.S. government got into jumpstations more than anyone else

As a concept, “jumpstation” long outlasted the search engine that popularized the term, even if it isn’t particularly a well-known term today. In fact, I know of at least one jumpstation that has been updated relatively recently.

The “Federal Acquisition Jumpstation,” a literal directory of directly of links regarding procurement and the purchase of goods, is a throwback to a time when long lists of curated links were the way that people found things online. Some people probably still do!

But what’s strange about this URL is not that it simply exists, but that the last time it was updated was just last year. The URL has maintained its current home since 2002 and I’ve found evidence that it was maintained at other URLs hosted on NASA servers since at least 1995. It has gone through at least three “owners” responsible for its maintenance within the executive branch over the years, and despite linking to pages that have changed significantly in the years since the Federal Acquisition Jumpstation was first launched, the design of the site itself has only gotten modest visual updates over the years. The result is that this may be one of the oldest periodically updated sites on the entire internet that largely exists in its original form.

And at one time, this jumpstation, which seems to have long outlived every other jumpstation, was seen as an important part of the way the government engaged with external vendors.

“The concept is to provide industry one stop shopping for all federal procurement information available through the Internet,” a 1995 NASA Procurement Countdown newsletter explained. “The Office of Federal Procurement Policy has been very supportive of this approach and will encourage all agencies to make their electronic data available through the Jumpstation.”

Unwittingly, this obscure resource about procurement speaks to the willingness of the federal government, given the right people and the right level of interest, to follow through on an idea even decades later—despite the fact that many potentially more useful ways to, uh, procure that information now exist.