How Gopher Nearly Won the Internet

By Scott Carlson | September 5, 2016 

[From https://archive.is/20211026235215/https://www.chronicle.com/article/how-gopher-nearly-won-the-internet/]

For a few years in the early 1990s, before the rise of the World
Wide Web, the internet was ruled by a humble computer program named
for a rodent. Gopher, created by a ragtag group of programmers at
the University of Minnesota, was the undisputed on-ramp to the
information superhighway.

The simple protocol that Mark P. McCahill, Farhad X. Anklesaria, and
their colleagues developed to transmit online information made them
the toast of the computer world. It was a brief but thrilling ride,
carrying them to conferences in Europe and across the United States,
where they met their idols in computing and spread the Gopher
gospel. In its heyday, Gopher demonstrated the potential of the
internet as an information system for the layman, beyond its status
as a tool (or toy) for tech enthusiasts, academics, and high-level
researchers.

But a mix of technological advances and commercial interests soon
drove the ascendant World Wide Web onto computers and into the lives
of pretty much everyone with an internet connection, effectively
burying Gopher. The ride for Mr. McCahill and his team would end
almost as quickly as it began.

Twenty-five years later, Mr. McCahill is wistful about the hit
program he helped create. "Like pop music," he says, software is
ephemeral.

"You could say that we were a pretty good grunge-rock band, but
grunge fell out of fashion, and what came after got a lot more
popular."

Gopher was born late in 1990 as an upstart alternative to a
long-planned information system that would serve the Twin Cities
campus. A campus-computing committee had been talking for years
about what such a system should do. So far the panel had compiled a
document thick with proposals and plans, yet no one had written a
line of code. Mr. McCahill, a senior university programmer, was sick
of it.

The committee's next meeting would be different. Mr. McCahill had
gathered a team of young programmers who handled computer-oriented
odd jobs and help-desk tasks for the university. Jamming to a
soundtrack of Nirvana, Mudhoney, and speed metal, they spent a month
of late nights pounding out a program that would allow even novice
computer users to get campus news, information about services, phone
numbers, email addresses, and so on. It was designed to run on the
PC - an emerging technology that posed a challenge to the dominance
of mainframe computers - relying on servers and the internet. At
that time, the internet was mainly the province of military
personnel, engineers, scientists, and computer enthusiasts, and
using it required some technical prowess.

Mr. McCahill and his team named their creation Gopher, after the
university's mascot. They had anticipated excitement when they
introduced it to the campus-computing committee, and excitement they
got - just not the kind they had hoped for.

"It was the worst meeting I have seen in my life," recalls Bob
Alberti, another member of the Gopher development team, who today
handles computer security at Minnesota. "People were jumping up and
down and screaming, ‘You can't do that!'"

The Gopher team had made a crucial miscalculation: The committee was
dominated by staff programmers who worshiped at the altar of the
venerable mainframe, and who saw, in the campuswide system, a shot
at salvation. "People had been fleeing mainframes like crazy" for
PCs and Macs, Mr. McCahill would say in a 2001 interview. Gopher's
surprise unveiling was received not only as a threat to the future
of the mainframe, but as an assault on the usual deliberative
academic process by a pack of "Mac/PC/Unix support monkeys."

"We pulled the legs out from under the committee," Mr. Alberti
says. Administrators told the upstarts to halt their work
immediately.

But it was the dawn of the modern internet era, where ideas spread
like a contagion. And this was, after all, a university campus, not
some top-down company where the CEO could order the genie stuffed
back in the bottle.

Mr. McCahill and his team of "PC radicals" - mainly Mr. Alberti,
Mr. Anklesaria, David Johnson, Paul Lindner, and Daniel Torrey -
continued quietly working on the program. They sent Gopher out to
colleagues at other universities and asked for help, which quickly
generated buzz about this new tool that could make finding
information on the internet easy for the layperson.

"We were a skunk-works project," says Mr. McCahill, "trying to find
allies anyplace we can."

The buzz put pressure on Minnesota administrators not only to adopt
Gopher as the campus information tool, but also to release it to the
world.

The mainframe team could never get its version of the campus
information system to work, Mr. McCahill says, "and working code
usually wins."

Gopher emerged at an opportune moment. Lots of colleges were trying
to create campuswide information systems, and Gopher made it easy:
Just set up a server, point to some of the information that the
Gopher guys had already put online, and you looked like a hero.

"It was the first viral app," says Mr. Alberti.

Long before Google's algorithms revolutionized web searches, Gopher
gave anyone with a PC and a dial-up connection access to the
collective knowledge and resources of the internet. Hungry for
seafood? You could hop on Gopher, type "salmon," and find
recipes. Want to play chess? Gopher could connect you with a game.

Perhaps you'd like to read about higher education? The Chronicle
became one of the first newspapers on the internet, in 1993, as a
Gopher service. Gopher would pioneer hypertext, the associative way
of linking related bits of information now common on the web, and
its search function paved the way for Yahoo and Google.

"If it weren't for Gopher, the web probably would have died," says
Charles R. Severance, a clinical associate professor of information
at the University of Michigan at Ann Arbor.

Mr. Severance, who teaches the history of the internet, says that
the World Wide Web, as it existed in 1993, was "a bad idea," not
merely a good idea that hadn't come to fruition. It required too
many resources for the limited technology of the time. Gopher
demonstrated a fast and simple way to distribute and access
information over the internet, and it stoked the energy for
information spaces among the "nerds," Mr. Severance says. But even
though Gopher would look primitive to us today, compared with the
free-form surfing of the web, he and his peers didn't view it that
way back in the day.

"We thought that we had seen rock 'n' roll at that point," he says.

Gopher was built on hierarchical menus, designed as a
straightforward way for people who aren't computer savvy to find
content on the internet. For example, the starting screen might
offer a menu of choices: information about Gopher, fun and games,
libraries, mailing lists, campus information, news, phone books, and
so on. Choosing one of those options might lead to other lists or
the resource you're looking for. Gopher mainly generated text,
although menu options could lead to pictures.

At the University of Minnesota, Gopher's creators kept their day
jobs as programmers. "The university never gave us any additional
resources to do anything with Gopher - everything we did with Gopher
was on the side," Mr. Alberti says. "I still had to do help-desk
support 20 hours a week."

Meanwhile, Gopher was blowing up the computer world. In 1992 its
enthusiasts organized a conference - the first GopherCon - to
discuss the future of the application. Mr. McCahill and his team got
to hang out with some of their computer-science heroes, like Alan
Kay and Jon Postel, whom some had dubbed the "god of the internet."
Even Adam Curry, the MTV video jockey, would at one point wear a
Gopher T-shirt on the air.

Mr. McCahill remembers a potentially pivotal moment during that
frenetic time: At a meeting of the Internet Engineering Task Force
in San Diego in 1992, a computer scientist named Tim Berners-Lee
approached him on the meeting's final day. "You know, I think we are
trying to solve the same problems here," said the Englishman, who
was working on an internet tool of his own, one that hadn't yet
gained traction. He proposed a collaboration with Mr. McCahill and
his team, maybe even merging their two projects. Mr. McCahill said
he'd consider it.

When he got back to Minneapolis, he read the documents but couldn't
follow the ideas or see how Gopher would align with
Mr. Berners-Lee's project. Mr. McCahill sent a note to the man - who
would, of course, become famous as the inventor of the World Wide
Web - wishing him well.

"I look back on that now," Mr. McCahill says, "and I think, You
know, that was probably a mistake."

Mr. Severance, from the University of Michigan, doesn't think
Mr. McCahill could have made any other choice. "Mark McCahill, until
it was over, was like Usain Bolt - he couldn't work fast enough," he
says. "We wanted so much of him. He was Plan A for the universe." If
Mr. McCahill had spent three months trying to merge Gopher with the
nascent web, "we would have been like, Stop it!"

In 1993, the University of Minnesota tried to charge a licensing fee
to use Gopher, and some people think that attempt at
commercialization led to its demise. But Mr. Alberti faults a
confluence of technological advancements in 1993 and '94,
particularly involving the transmission of images. Modem speeds
spiked - which meant that people could download images faster on
what was known as the "World Wide Wait." Marc Andreessen came up
with the "img" tag that would allow someone to put a picture on an
HTML page, and his Netscape Navigator could display a web page's
text as the images were downloading. On Gopher a user had to go to a
separate file to see images.

"Who is going to look at a bunch of text lines when you can look at
a photo of Taylor Swift?" Mr. Alberti asks. The image capability
also appealed to the vanity of internet users, who wasted no time
creating personal web pages to promote themselves.

Mr. Severance recalls an anecdote, perhaps apocryphal, to illustrate
how fast people migrated to the new technology: At an IETF meeting
in Columbus in early 1993, a session on Gopher was packed, while a
despondent Tim Berners-Lee spoke to a nearly empty room. But by the
end of 1994, computers were being shipped with pre-loaded
web-browsing software, and Gopher's dominance was over. Bit by bit,
its servers shut down, and Gopher went underground.

Today, a few tech nostalgists keep Gopher alive - and they are not
persuaded that the protocol is obsolete. One prominent advocate is
Cameron Kaiser, a computer programmer and physician who runs the
Overbite Project, an effort to "return to the Gopher protocol's low
bandwidth and high efficiency" by providing the software for modern
devices and computers. In a manifesto on one of his web pages,
Dr. Kaiser praises Gopher's straightforward, reliable operation and
its purity as an information space, free from commercialization.

"Gopher sites stand and shine on the strength of their content and
not the glitz of their bling," he writes.

Peter Garner, a data-communications analyst in Halifax, West
Yorkshire, in England, started using Gopher two years ago when he
wanted to save energy by slimming down the servers he operates in
his home. He mainly uses them to distribute political documents and
local news - "text files that go a little bit beneath the web, but
not quite to the darknet," he says.

Gopher still has powerful potential and shouldn't be relegated to an
"anorak's plaything," he says, a nod to British geeks who wear
anorak jackets. "In developing nations," Mr. Garner says, "I think
Gopher has lots of potential because it literally can run on dial-up
with minimal hardware. What more could you want?"

In much of the Western world and beyond, what people want is
immediate images of important events from across the globe - along
with, of course, cat videos, Amazon Prime, and pornography. All of
that is readily served by the World Wide Web.

Mr. McCahill recently downloaded a version of Gopher for his
iPhone. "Doing that reminded me of what things used to be like, when
there wasn't an ad on everything you went to."

There might be a place for Gopher, even in a web-dominated world,
but Mr. McCahill wonders if today's computer nerds would bother
rediscovering it. Unfortunately, he says, few computer scientists
know their field's history, so they keep reinventing things.

Mr. McCahill was sad when it became clear that the web, not Gopher,
would dominate the internet, but he says that Gopher enriched his
life, if not his pockets. He left for Duke University in 2007. The
University of Minnesota subsequently awarded him a place on its
"Scholars Walk" alongside Minnesota luminaries like Bob Dylan,
Patricia Hampl, Hubert Humphrey, and Garrison Keillor.

The most enduring gift of the Gopher experience, Mr. McCahill says,
was a humble lesson: "I understood that it was OK to do a simple
thing," he says. "You didn't have to do something
complicated. Simple could be very powerful."