SLAC Symposium on the Early Web
Dale Dougherty and I have been invited to talk about our experiences creating GNN, the first Web portal. Since we sold GNN to AOL in 1995, and they shuttered it less than a year later, its place in early Web history is now known mainly to those who were there at the time. (Fortunately, John Labovitz, one of our developers, kept a snapshot of an early version of the home page.) GNN, or the Global Network Navigator, created in the fall of 1992 and launched early in 1993, was the first purely commercial Web site. It was the first Web catalog (predating Yahoo, and incorporating the NCSA What's New page), the first Web-zine, and it introduced advertising to the Web.
Our first ads were simply buttons pointing to commercial "resource centers," and the actual ads were pages that we hosted on our own site, since there were not enough people on the Web yet to point to other sites. (By early 1994, though, we did have ads on content pages, as indicated in this archived mailing list posting.) The very first advertiser on the World Wide Web, by the way, was our Silicon Valley law firm, Heller Ehrman White & McAuliffe. Needless to say, they now have their own Web site. But it was an uphill battle at the time to get anyone to consider the possibilities in the new medium. Not only was it hard to convince potential advertisers that the web would work for advertising, it was going against the prevailing ethos of the R&D-focused internet to have any kind of commercial activity. (For a small anecdote about the National Science Foundation and the crumbling of its non-commercial only Acceptable Use Policy, see the postscript to this message I posted to the Studio B mailing list last year.)
GNN started out as an online version of the catalog portion of our book, The Whole Internet User's Guide and Catalog. (This book is out of print now, but a whole generation learned of the Internet from its pages. It sold over one million copies, and was selected by the New York Public Library as one of the most important books of the 20th century.)
Our goal in the catalog was to highlight the best sites on the Internet. In those days, Web sites led to only a tiny proportion of the resources available, rather than the dominant gateway they are today. Instead, you might telnet to a library catalog, ftp to a file download site, or use WAIS or Archie to search for sites. The catalog gave instructions on how to reach the top two or three hundred services. Pei Wei, who was working on an early Web browser called Viola, showed us how he could use the Web to make a point-and-click way to reach all these resources, so people wouldn't need to type in long, obscure command lines. We were talking about setting up Internet kiosks in key technical bookstores to help promote the book by promoting the Internet itself. Pei said "I can make a cool demo for you." But when he showed it to us, Dale and I said, "That isn't a demo. That's a product." (The Whole Internet kiosk was actually deployed briefly at Computer Literacy Bookshop late 1992, but it was never rolled out to more bookstores, since we decided to pursue the GNN portal idea instead.)
If we'd been software developers, Viola might have become Mosaic, and Pei, Marc Andreesen. (Mosaic was inspired by Viola, which was really the first graphical Web browser. Pei also may have been the first to make "view source" a key part of the web experience, at least according to this 1992 review of Viola by Tim Berners-Lee.) But we were publishers, and so we were entranced by the possibilities of Web publishing. Dale wanted to start a Web magazine. I wanted to know how we'd get people to pay for it, and had a brainstorm the day my copy of InfoWorld arrived. There are lots of free magazines supported by advertising, I said. We could do that online as well!
Another key debate was the frequency of publication. At first, we thought it might appear quarterly, then monthly, then weekly. Finally, we said, "People are going to have their Web browsers open on their desks all day!". In late 1992, that was an ambitious vision, but it turned out to be more than true. (In a survey we did in mid-1994, several years later, when the revolution was further advanced, 77% of GNN users said they were online once a day; another 8% said several times a week. Only 13% said they were online more than once a day. The median online usage was seven hours per week. So our vision that the web would be ubiquitous and constant was more of a stretch than it might appear now.)
When we released GNN in early 1993, people didn't know what to make of it. The now universally understood metaphor of the Web site framed by the browser was not yet common. What was the software, and what was the content? What's more, how did you get this wonderful new thing? There was a long answer, and a short one. The long answer involved a whole book's worth of instructions on getting onto the Internet (there weren't many ISPs yet), downloading a bunch of software, and then pointing to GNN. We started looking for a simpler answer.
Working together with a Seattle software developer called Spry, we developed a product called Internet in a Box. This was a way to simplify the distribution of GNN. Our ideal was to have the phone company just offer the Web as part of their service, as France Telecom did with Minitel, but despite many approaches to the Baby Bells, no one got the idea till much later. The proprietary online services were interested, though. Spry sold Internet in a Box to CompuServe, and we sold GNN to America Online, and neither pursued our vision till too late, leaving the Web portal market to Yahoo and others.
Meanwhile, we continued with Web publishing. Our idea, outlined in my 1995 paper, Publishing Models for Internet Commerce, was that the industry would evolve into a multiplayer model, with Web authors and publishers, but also Web "wholesalers", if you like, sites that aggregated and distributed user attention. This has turned out to be the case, with sites like yahoo.com, aol.com, msn.com, and google.com directing a huge proportion of Web traffic.
As to advertising, we started out with the hope that the Web would lead to a new era of content-rich advertising instead of quick jingles with little or no information content. Despite the proliferation of increasingly intrusive advertising, starting with rotating ad banners up through today's pop-ups, pop-unders, and streaming media downloads, there was a way in which we were right. All those intrusive ads are really just advertisements for the real ads--the sites themselves, where rich product information can actually be found. For example, oreilly.com provides far more detailed information about our products than could ever be found before the era of the Web.
The Web has come a long way in the past decade. But what's exciting is that it has even further still to go.
Tim O'Reilly is the founder and CEO of O'Reilly Media, Inc., thought by many to be the best computer book publisher in the world. In addition to Foo Camps ("Friends of O'Reilly" Camps, which gave rise to the "un-conference" movement), O'Reilly Media also hosts conferences on technology topics, including the Web 2.0 Summit, the Web 2.0 Expo, the O'Reilly Open Source Convention, the Gov 2.0 Summit, and the Gov 2.0 Expo. Tim's blog, the O'Reilly Radar, "watches the alpha geeks" to determine emerging technology trends, and serves as a platform for advocacy about issues of importance to the technical community. Tim's long-term vision for his company is to change the world by spreading the knowledge of innovators. In addition to O'Reilly Media, Tim is a founder of Safari Books Online, a pioneering subscription service for accessing books online, and O'Reilly AlphaTech Ventures, an early-stage venture firm.
Comments on this weblog
Return to weblogs.oreilly.com.