I was there at the birth of the web, and I’d like to share my story…
(I’m not talking about the awesome sprouting of the underpinnings of the World Wide Web in the early ’90’s at physics research hubs like CERN and SLAC, but rather the Big Bang of the commercial web that rapidly emerged upon that foundation in 1994 and 1995 at a bunch of startups across Silicon Valley, unleashing one of the biggest waves of game-changing entrepreneurship the world has ever seen.)
I was at the veritable right place at the right time for this once-in-a-career opportunity, having become the product manager for the Indy workstation at Silicon Graphics (SGI) in early 1994. That bright blue UNIX “pizza box” (which, painfully, I admit most of you have probably never heard of) was truly at the center of the action at the earliest days of the commercial web.
You see, the Indy workstation was the development platform for Marc Andreessen and Eric Bina at NCSA when they created the Mosaic browser. And it became the primary software development and web-serving platform in the earliest days of Mosaic Communications Corporation (an electrifying startup, founded by Marc and SGI’s founder and recently departed Chairman, Jim Clark, that would soon change its name to Netscape). And it was the sexy server hardware proudly used by some of the most prominent sites of the early web, including HotWired (the first online magazine), Organic Online (the very first interactive ad agency), and Virtual Vineyards (not only the first online wine seller, but the very first web-based retailer, period). As Indy’s product manager, I had a unique opportunity to follow my product into one of the most rapidly exploding markets in the history of computing — an opportunity I seized with gusto.
But as it turns out, those very early days of the commercial web, long before the “dot com bubble” of 1998 and 1999, are, for the most part, ironically and tragically ungoogleable. Of course, the Way Back Machine gives some glimmers of the old days, but its earliest records go no further back than 1996. And while Wikipedia has posts that give some of the backstory, the fact is that very little remains of the websites, press releases, and news stories of 1994 and 1995.
So, I believe that now is a good time for me to start writing down my memories of that historic time. With luck, it might inspire others to come forward with their own anecdotes.
In the coming weeks, I plan to share first-hand accounts of the inside stories behind a number of industry “firsts,” including the first advertising deal of the web, the first business-oriented web conference, the first platform-wide licensing deal for Netscape, the first visual HTML editor, the first web server product line, and the first licensing deal for Java. Back then, we cared a lot about what print publications wrote about us, so I hope to include some photos of long lost pubs, like Interactive Week. 🙂
If you were a part of those stories and want to add to the narrative, please email me or add your comments along the way!
The Real McCrea 
The first DNA Molecular Core Web Site.
I came to the Joslin Diabetes Center in August 1994 to set up and run a new DNA Molecular Core Facility. The facility would sequence and synthesize DNA and perform Realtime PCR. We used ABI equipment which came with Apple Macintosh computers running ABI software. I had previously worked with the equipment and software in Children’s Hospital Boston since 1991. I wrote my Zoology thesis on a Mac in 1989. So at the time, I was definitely a Mac guy.
Now that I was running my own facility, I wanted things done my way. I was taking requests from the researchers on paper forms that I created and printed. This was a very inefficient way to file and keep track of records. I needed the process computerized, I needed a database. But I didn’t want to transcribe the information from paper forms that was filled out with pen by a user. That meant moving the information twice. What I really wanted is for the users to put the information directly into a database themselves with no need for hands on by me. But how? Put a computer in a shared space and let users sit at it and enter their requests into the database? Bad idea. Someone is going bust the database or delete it. How about a web site with a back-end database? A what?
I went to the IT department guys, back in those days academic IT guys really didn’t have any sense of urgency. I told IT that I wanted them to build me a web site that would have forms and allow users to submit DNA sequencing requests, insert the information into a database and send a confirmation email to the user and one to me. Not surprisingly they laughed at me but I could detect that sense of “We don’t have a clue what you’re talking about or how to do that!”
I did some research and found a Mac web server called WebStar, an add-on named NetCloak for Server Side Includes and FileMaker Pro for a database. I learned some HTML and built my web pages and forms. I also learned some AppleScript and wrote CGIs to process the forms and send emails.
I threw all the paper forms in the recycling bin, got a FQDN for the site under harvard.edu and spread the word to all the Joslin researchers that they needed to start using the web site for all their requests. As well as web forms for submitting requests, I created web pages with all kinds of content describing the process and science behind what we did in the facility. An interactive web site with a wealth of information.
The Joslin Diabetes Center published its official web site about 3 or 4 years later in the late 90s. In the meantime the Joslin Diabetes Center Molecular Core Facility web site was in full swing. We received requests from other hospitals and pharmaceutical companies both in the Boston area and beyond. We even had requests from the UK and Mexico, all because they found us through a web search.
For me, this was the early days of the Web. I wanted to be part of it and did on a Mac, the only way I knew how. I left Joslin in 2000 and soon afterwards they shut down the lab and the web site with it.
Thanks for sharing your story from the early days of the web!
I did some cool things with that blue pizza box.
Thanks for that, John. I was at Cray Research in those days, and then SGI, and then Cray again — exciting times. It is sad that pre-Web history has to be painstakingly recreated and imported piecemeal onto the Web. History on the Web is still mostly a blank slate for the era before the mid-1990s.
It’s too bad that SGI did not see the opportunity to leverage the Indy in web hosting
in those days. We made several proposals for appliances, web, gateway (security) and NAS filer and all were shot-down by the server BU. The GM later told me, several years later, that it was the worse decision he ever made. So much for advanced thinking at SGI
in those days. No wonder why Jim Clark left to do Netscape.
Well, not to get too far ahead of the story, but we did end up leveraging the Indy (and the Challenge server line) for web serving under the WebFORCE brand. Jim left and re-connected with Marc Andreessen not much thought about the web. In fact, they were more focused on video games, then shifted gears to the web.
The Challenge-S line (Indy w/o graphics) was very successful and led the way to the aborted Internet Server Environment series of web/email/calendar/e-commerce servers. I remember SGI asking some porn companies to remove the “Powered by SGI” logo on their web sites.
John,
Sure does seem like a lifetime ago working on the WebForce rollout. Will always be some very fond memories for me.
[…] minutes something caught my eye. Indy had a very distinctive bright turquoise shell. But sitting on a shelf a bit to the side of the assembly line was something that looked just like […]