The Father of VoIP Moves On

Richard Platt

Chances are good that you don’t know who Richard Platt is. Richard was never in the business of self-promotion - had he been, he could have been a household name among the cognoscenti. But Richard didn’t spend his time talking to the press, speaking at conferences, or teeing up with the titans of industry. His passion was devoted to building great products and taking care of the people whom he asked to share that passion.

A Short History of VoIP

In early 1997, David Tucker and Richard Platt were running a division of Intecom called Incite. They were developing and promoting an innovative product designed to optimize the use of video on corporate networks, using a backbone of isoEthernet. I spent some time contracting for Incite, and believe me, in 1997 it was pretty cool to be able to watch four channels of broadcast TV on a PC. But the technology that made this possible, isoEthernet, was also the weakness of the Incite solution. Interface cards were expensive, and isoEthernet wasn’t compatible with emerging Fast Ethernet standards.

David Tucker

This could have been the end of the story, one that is played out daily in technology companies - a bet that didn’t pay off. But instead, a couple of remarkable things happened. First, Richard and David sat down and decided that their existing call processing software and engineering expertise could take a U-turn and instead produce a Voice PBX that ran over the network - all signaling and media would travel over the corporate network instead of using a separate wiring system. The software that had been routing multimedia traffic over isoEthernet would have to now route voice calls over IP, and a complete set of apps and hardware would be needed to complete the effort.

A willingness to scrap a big project and divert a skilled engineering team to a completely new venture is unusual enough. The second amazing thing that happened was that Richard and David were able to convince Intecom management that this was a viable scheme. Company President George Platt (no relation) gave the concept a green light, and suddenly the pressure was all on Richard’s shoulders. Richard’s engineering team had to show that the concept of network-based telephony was viable. There were no commercial VoIP systems at the time, and in fact there was a lot of skepticism that the concept could get off the ground. The pressure was on.

Engineering Crunch Time

Paul Hahn

What ensued was classic Richard Platt. Having successfully made his pitch to management, he now had to get his Engineering team to produce a completely new product from a standing start - and time was of the essence.

Fortunately, Richard has an asset that he shares with other great technical leaders: a high-powered Reality Distortion Field. Richard told his engineering managers what had to be done, and his belief in their ability to produce infected them with the confidence they needed to do the job. (I’ve seen this again and again with Richard. His belief in the people who work for him is so sincere and so infectious that people respond with herculean effort to make his visions come to life.)

Before the year was out, Paul Hahn’s team had hacked a basic Intecom phone into a VoIP network appliance, John Alexander’s team had created the call processing software needed to perform basic PBX functions, and Jeff Sanders’ team had a soft-phone and initial set of applications ready to go.

The system was debuted to Industry Pundit Jeff Pulver in October 1997, perhaps marking the first public demonstration of a commercially ready VoIP phone system. It was followed a few months later by a demonstration at the PBX 2000 show in Washington CD, and the debut of a new company, Selsius Systems, Inc. Despite claims to the contrary by 3Com, Selsius Systems was the first company delivering a scalable PBX that ran over the network¹.

The Next Level

John Alexander

Being first to market is a great honor, but it doesn’t give you a lock on success. Witness Netscape, the Apple Newton, eToys, etc. In their first year of selling systems, Selsius was able to move a few thousand phones, but mostly to tire-kickers. The big enterprise buyers weren’t going to buy a new technology from an unknown company.

At this point David and Richard again worked some magic - convincing Intecom that it was in their best interests to sell Selsius to a big, established company that could give the technology the sales push that it needed. Cisco Systems stepped in with a good offer and took the deal away from runner-up Northern Telecom. (Given the astonishing fall in Nortel’s fortunes over the next few years, this decision seems charmed in hindsight. At the time it was not so obvious that Cisco, with little telephony experience, was a better partner for this new business.)

Without the Selsius acquisition, it is very unlikely that Cisco could have become the dominant player in VoIP in just a few short years. Although our Cisco badges say we have “No Technology Religion,” the truth of the matter is that the Selsius CallManager architecture could never have been built in-house at Cisco.

CallManager was a Visual C++ call control app running on Microsoft servers, using an Access database, and communicating with telephones using the proprietary SCCP protocol. While these decisions were all perfect for the product, they just wouldn’t have happened inside Cisco engineering.

Fortunately, when the Selsius acquisition was complete in late 1998, the rest of Cisco Engineering just had to swallow hard and make the best of it. And make the best of it they did. A huge effort was launched, involving hardware engineering, sales, marketing, and software. As this progressed, Richard took on progressively greater roles in Cisco Voice, climbing the ladder to the lofty twin perches of VP and GM of the Voice Group.

Jeff Sanders

After just a few years of hard work, Cisco’s CallManager became a billion dollar business, with at least that much in additional pull-through for other Cisco hardware. Cisco now ships almost half the IP phones lines (see Infonetics) in the world, and as people move inexorably to VoIP, their total share of the entire voice market will only increase. CallManager has beefed up to a distributed system that runs on Linux or Windows servers, and can talk to SCCP or SIP phones. Systems can support 30,000 users in a single cluster. TV viewers know that even US President Bartlett uses Cisco IP phones, along with his crack anti-terrorism team at CTU, including Jack Bauer.

Five Years at the Helm

After guiding Cisco’s voice products to this level of success, the entrepreneur in Richard started to get itchy for a new venture. With his track record, he could have easily raised money to start a new company, but instead he decided to try to create a startup within Cisco.

Richard and David moved to Linksys in 2005 with the goal of creating a competitive small business VoIP product. Like in the Selsius days, this required some deviation from standard Cisco engineering practices, and Linksys seemed like a safe harbor.

The resulting product, the Linksys One product line, is just now achieving a level of maturity that is allowing it to compete in businesses much smaller than those that CallManager can service. Will it be the home run that CallManager was for Cisco? It’s too early to tell, but with ongoing commitment from Linksys and Cisco management it certainly has a fighting chance.

Coda

At the end of May, Richard started circulating the word that he was retiring from Cisco. There was much wailing and gnashing of teeth from his devoted followers, but at the same time, most of us understood his reasons for leaving. If you love engineering, creating products, and working with people, life as a Vice President at Cisco combines incredible demands and frustration with little joy. It’s not an easy job, and it is often not a fun job. In fact, general consensus among those who knew Richard well was surprise that he lasted as long as he did.

And so in June, after the usual round of farewell celebrations, souvenir gifts, and a few tears, Richard was gone. He was coy about his future plans, saying right now he’s just going to relax and spend some time in that time-honored pursuit of the well-heeled, getting his private pilot license. It may be that a no-compete agreement with Cisco has something to do with that storyline. Or it may be that Richard’s life outside of work is full enough to keep him engaged the rest of his life. Time will tell.

Richard, David, and the rest of 30 or 40 original Selsius team deserve credit for launching an industry, and for creating a huge new business for Cisco. You can argue that VoIP was in the air, was inevitable, and that somebody was going to do it if Selsius didn’t. That’s true, but Selsius was not only first, they managed to turn that first mover advantage into total dominance through hard work and smart business. Richard was a huge part of that, and Cisco is going to miss him.

As will I.

Footnote

1) The truth about who delivered the first commercial VoIP PBX may never be truly resolved. 3Com and Selsius unveiled their solutions nearly simultaneously. One of the biggest problems with 3Com’ s initial release was that their system was limited to 255 phones, which was a problem when trying to sell to the enterprise. 3Com’s product is still available, and has seen a great deal of enhancement, but Cisco’s giant gravitational field sucked up so much market- and mind-share that 3Com has had to settle for a role as niche player.

If you want to make the case for 3Com’s role as #1, and have some good facts, by all means please add them to the comments section!