It was the month of November 1994, and my mission was clear: create and launch the first hardware + software web product line, putting Silicon Graphics (SGI) into the leadership position of a market about to explode. Based on what I learned in Chicago, I believed that if we launched by end of January, we had a very good chance of beating key rivals Sun and Apple to market with solutions for the most valuable “picks and shovels” segments: web authoring and web serving. To get there, I had a two-part plan, focused on forging internal parternships with two other divisions within the company.
The first part was to partner with NSD (Networked Systems Division), a business unit that had been formed within the last year to expand the company’s footprint beyond 3D workstations into the market for high-performance servers. If our two divisions worked together, we could create a product line with the Indy workstation as one entry point (a multimedia monster for both web authoring and web serving) and the Challenge S as the other entry point (NSD’s Indy-based entry-level offering, an ideal web server). The rest of the Challenge server line (see image below) would provide a path for massive scalability.
The second part was to partner with one of our two software divisions to create an awesome, SGI-native, graphics- and media-centric, drag-and-drop, “WYSIWYG1” web authoring tool that we could bundle with our workstations.
Since it was more than a little ambitious to dream of creating such a tool in less than three months, I started with the authoring software. My first stop was Silicon Studio, a division recently formed to create authoring software for interactive content. Heading up product management there was none other than Scott Bonham, who had hired me into the company and then given me a big break by moving and an allowing me to become his backfill. Despite my clarity, conviction, and passion, nothing I said could convince him that the web should be anything other than a secondary focus for Silicon Studio. Their plate was already full, working on an authoring system for video games.
Oh, well. At least I had one other software division I could partner with, VMD (Visual Magic Division), the group responsible for SGI’s “Indigo Magic” desktop environment (see image below) and a bunch of cool software tools highlighting unique capabilities of our systems. They even had a team focused on publishing. But conversations there fared no better. My logical champion to take on the project was dismissive of HTML, seeing it as an inferior derivative of SGML2. He would happily lead an effort to build an SGML editor. I thanked him, but suggested the web was unlikely to switch to SGML, even if we built the best damn SGML authoring system in the world!
Suddenly, SoftQuad’s HotMeTaL Pro3 started to look much better to me, despite its unlovable interface. I set to work negotiating the terms of a deal to get it ported to IRIX, so that we could bundle it with the workstation part of the web product line I planned to launch in January.
And To Serve
I hoped I’d fare better with the server division. After all, the plan there required no software development, just a simple bundling of existing hardware units with third-party web server software (to be licensed from Mosaic Communications). SGI would be first to market with a “turnkey web server” for very little effort. Lenny Rosenthal, who headed up marketing for NSD, was conflicted. He saw the great potential of the market, but knew that convincing his boss, Ross Bott, the GM of the division, would not be easy. He arranged a meeting on a Thursday afternoon for us to make the case. Ross was a seasoned hardware exec, very technical, and methodical. I, on the other hand, was inexperienced, learning Computer Science on the job, and given to a passion that might be called “evangelical”. The meeting clearly did not go well, but no decision was made on the spot. The next day, however, Lenny told me that NSD would not be jumping with me into the web server market. Instead, they would stay focused on the market for high-scale database servers.
Dead in the Water
My bold plan could survive without homegrown web authoring software. But trying to make SGI the leader in the web server market without the participation of our server division? That was laughable. Our first-to-market web product line was dead in the water.
And that was simply not acceptable.
It was a Friday afternoon (likely November 11). I went to see my boss, Jim White. Readers of this series may recall that Jim had taken over the head of marketing role for the division behind Indy, DSS (Digital Sight and Sound), and immediately empowered me to focus my energy on finding new markets. I shared with him my frustration with the disastrous results of my internal partnering efforts, concluding with, “SGI is in the perfect spot to take advantage of the biggest new market in computing, and we’re going to totally blow it. We’re going to just let it slip through our fingers.”
Jim’s a pretty intense guy. Smart, confident, and passionate. He didn’t hesitate in his response, as if what to do should have been completely obvious to me. He asked, “Why don’t you just do it?”
The light bulb went off. It hadn’t occurred to me that such an option existed. But as soon as he said it, I realized that there was a pathway forward, if I could grab full leadership of a company-wide effort. “I’m in,” I said.
Minutes later, we were in the cubicle of Tom Furlong, General Manager of DSS, pitching the idea of creating a new product line by bundling software for web authoring and web serving. Tom, who embodied the epitome of the the can-do spirit of mid-’90’s Silicon Graphics, was immediately supportive, saying that he had been advocating for “solution bundles” for a while. “Let’s do it,” he said. “What do you need from me?”
“I need to be knighted,” I said.
Tom smiled, and without missing a beat, pretended to pick up a sword. “You are hereby knighted,” he said, dropping the virtual blade to my shoulder.
“That’s awesome, and I really appreciate it, but I think this knighting will have to come from higher up,” I said. “This is a mission that requires product changes in multiple divisions.”
Still smiling, Tom stepped toward his desk and picked up the phone. “Can you be ready to present at TJ’s business ops meeting on Monday?” (“TJ” was Tom Jurmoluk, the company’s COO.)
Without hesitation, my mouth said, “Yes,” but my mind thought, “Whoa, that’s crazy. I don’t even have a draft presentation!”
A minute later, Tom hung up the phone. “Okay, Monday’s totally booked, but you’re now on the agenda for the exec team meeting on Wednesday.”
I thanked Tom, and as we headed out of his cubicle, he said, “One thing, though. I don’t want you go hat in hand, asking for money.”
I nodded and quickly left, eager to start working on what might be the most important presentation of my Silicon Valley career.
To be continued…
1This is an old acronym for “What You See Is What You Get” which came into common use in the 1980’s during the word processing revolution. From Wikipedia “a WYSIWYG editor is a system in which content (text and graphics) onscreen during editing appears in a form closely corresponding to its appearance when printed or displayed as a finished product, which might be a printed document, web page, or slide presentation.”
2“Standard General Markup Language,” the father of HTML.
3The first commercial HTML editor, which I had encountered weeks before.