NTI:1994-1995

So, I had this idea in April 1994 in the shower of 705 Lakeview Way.  I thought to myself, “What if I could build a box that would change the IP addresses that a network has already had configured into a set of registered IP addresses that could be used on the Internet?”  Part of the engineering work I was doing for network analysis involved writing code using the Sun NIT (Network Interface Tap) which would pull the whole packet off the Ethernet and present it to the upper layers of software.  Perfect!  This would be a way to get the packets I want so I could modify them properly.  So, I then got on the phone with Brantley Coile in Athens, GA (see image of phone bill), the best engineer I had ever known on April 5 to tell him about this idea.  I went over all the details with him.  He then came back to me with a game changing idea for this product.  He suggested that we develop our own “real time executive” to do all the work of getting packets, this way we weren’t a slave to any of the bugs in the SunOS.  I was at first a little hesitant because I knew it would involve more work and time, but in the end, we discussed it and I got more comfortable with the idea.  Before Brantley got off the phone, he yelled to his wife, “Hey Betsy, I think there is a new house for us with this idea!”

So, JMA was going great guns and Johnson and I were working non-stop.  Brantley was consulting for a few other customers in Georgia, so I became another consulting customer of his as we did the long distance development effort.  I would draw designs for how the system should operate, he would write the code, then I (and later Johnson) would test the code to be sure it worked the way it was supposed to.  This went on for many months in 1994.  JMA brought in all the money to build the company and the product (we had no investors) and we all worked together to make it a really solid system.

Brantley had to finish up a few other projects, and if memory serves me, he started on our product around April or May 1994.  I wrote him the first check on July 21, 1994 for the first batch of code.  By this time, he was full time on the “Translator Box” as we called it at the time, and was working very long days.  The “Translator Box”, was initially designed to be a box to provide access and change IP addresses.  As I went through the different scenarios for how it could be installed at a customer site, it was clear that it had to be a firewall along with a translation device.  So, I examined the router configuration that JMA used for most customers and converted that into a diagram that showed how to allow and deny access from the inside and outside for various protocols.  I asked Brantley to add these features to our product so that the Translator Box could become a firewall as well.  As a nod to the great company where Brantley and I worked together in 1989, we eventually called this security implementation “Adaptive Security”.

As usual for me, I chose a bad name for the company and the product to start.  To add insult to injury, I also chose to incorporate in Nevada to try and save income tax for the company.  So, we started out on August 31, 1994 as a Nevada Corporation, named Mesa Networks, with a product called IP Passport.  You can see our first attempt at a marketing plan here.  Sometime in 4th Quarter 1994, Andrew Heller joined the company as our Chairman.  The board always consisted of Andrew and myself.  When he joined, he insisted that I go speak with Richard Char at Wilson, Sonsini, Goodrich and Rosati (WSGR).  WSGR at the time was the premier law firm in Silicon Valley for high tech firms and Richard Char was a very smart attorney.  He and Andrew told me very clearly: Focus on running the company and making money, don’t worry about taxes and such now.  So, I filed for the trademark Network Translation in November of 1994 and by January 1995, we were Network Translation, Inc, a California Corporation.  While working with Richard Char at WSGR, I explained to him how our product worked like a PBX (Private Branch eXchange for telephones) for the Internet.  It allows for a small set of IP addresses (telephone lines) to be used for a large number of computers (phones).  He knew we were struggling for a name for the device and while he was explaining it to his paralegal, he said it is a Private Internet Exchange, and thus was born the name PIX.  You can see the trademark application filed in December 1994 here.  But I am getting ahead of myself with the business side of the story...

In November of 1994, one of JMA’s big customers, Alan Helbush of KLA Instruments, had agreed to be a beta site for our product.  To my recollection, they had 1,000-2,000 computers on their network and most of them had unregistered IP addresses.  We installed the PIX on their network and worked with KLA to get their computers aimed at the PIX to access the Internet.  The hard work of all of us paid off now.  The PIX worked and after a few updates, on December 21, 1994, Alan Helbush of KLA Instruments signed off on a successful beta test and gave us our first check for our product!

In parallel with all this, we were hitting the marketing trail very hard.  I did a tour of all the press on the East Coast.  Each year, the magazine Data Communications does a “Hot Products” award.  All the work of explaining the operation of the PIX (a totally new concept at the time) to all these magazines paid off.  We won the 1995 Data Communications magazine Hot Product of the year award for the PIX!  With this notoriety, the phone started ringing from customers and resellers around the world.  About the same time (the end of 1994), Shannon McElyea (who knows EVERYONE in The Valley) asked me if I wanted to be introduced to 3Com (Eric Benhamou) or Cisco (Ed Kozel).  I picked Cisco.  Shannon took me as her date to Ed Kozel’s Christmas party in 1994.  Unbeknownst to me, towards the end of the party, Shannon told Ed her date had created a network address translation device to solve network security problems and the shortage of IP addresses.  Ed had dozens of guests, but for some reason, he sat down with me for about an hour and half to talk about this new device we had built.  Shannon helped me forge the relationship that would end in one of Cisco’s most successful acquisitions of all time.  Shannon also showed me how to work with resellers and distributors and would be in the office at all hours of the day and night to communicate with them.  I was just an engineer trying to be a CEO.  Shannon helped all of us in countless ways to go from a small consulting group to a real company.

In early 1995 we hired more people at NTI.  We hired Rebecca Jepsen as VP of Sales and Susan Van Gelder as a sales assistant to Rebecca and general help in the office.  Between Rebecca, myself and all of our resellers, we managed to increase our sales each month.  Back then, our purchase orders would come in by FAX.  There was a bell like you see at the front desk of a hotel next to the FAX machine.  When a P.O. came in, whoever got the FAX rang the bell.  The bell rang more and more each month.  We ran very short of money on two occasions in the first 6 months of shipping, but by June of 1995, we were finally turning a profit.  In our last 3 months of operation as NTI, we had sales of about $1,200,000.00 (see sales figures here) with our tiny staff of full time employees: John, Richard, Johnson, Andrew, Rebecca, Susan, Brantley as a full time consultant, and a few part-time consultants.

In the Summer of 1995, Mike Volpi from Cisco had come to visit us and offer an investment in our company.  He and I worked on this for a few months until Ed Kozel (then CTO of Cisco) returned from Austria.  Ed sent over one of Cisco’s original and best Engineers, Jim Forster, to have a look at our product and our source code.  After he saw that we had crafted each line of code ourselves, he remarked with a big smile, “This is the way it used to be at Cisco”.  Ed took me to lunch at a restaurant called the Lion and Compass in Sunnyvale to talk about things.  After a few updates, he directly asked, “Would you like to sell your company to Cisco?”  I asked how it all worked and told him to come up with an offer.  He did, we negotiated, and I finally agreed to the terms.  The law firm of Brobeck, Pleger and Harrison drew up the merger documents which we signed in October of 1995.  We closed the deal in November 1995 and officially became a part of Cisco.

To be improved and continued...