FiOS: We start talking about GPON

Well, this is going to be a bit convoluted due to all the things that went on at the same time. I am going to finish up today with a bit more about Broadlight. Next week, I talk about Ethernet and the UMC. That sets the groundwork for the disaster that was the GPON RFP from Verizon.

Okay, we began to hear rumblings about a move to GPON from Verizon. Nothing formal, but questions and off hand comments. So, we started doing some math inside Petaluma. In particular, we worked with what we thought was the worst case for bandwidth - which was HD IP Video Streaming. We concluded that if this was where Verizon was headed then we had to move straight to 2.4/1.2 Gbps GPON.

We restarted our look at GPON proof of concepts (remember Tellabs had cancelled the ongoing AFC project at the time of purchase). And we started by chatting with Broadlight.

The relationship was a bit good and bad at this point. AFC had designed in 3 pieces from Broadlight: The OLT ASIC, OLT Transceiver, and ONT ASIC. Now this had changed over time. AFC designed out the OLT Transceiver and replaced it with a part from NEC. We were having issues with quality and scaling of these parts and wanted to go with a more substantial vendor. When we shifted our business to NEC the President of NEC Optical showed up and asked me and John Schofield "Why did you buy your entire order at once?" We answered, "I don't think you understand. We are placing the exact same order again next week." So scaling and volume here were challenges for everyone.

When Tellabs bought Vinci, we moved to Vinci ONTs after we made them comply to all the Verizon requirements. These products did not use the Broadlight ASIC.

So by the end of 2005, Broadlight went from owning 3 sockets to owning 1 socket. But now Tellabs was still the single largest customer and an investor.

So, we presented the opportunity of GPON migration to Broadlight and our desire to move straight to 2.4 Gbps. The response was that Broadlight was going to do a 1.2 Gbps Broadlight solution. We pressed very hard on this and Broadlight didn't move. I understand from their perspective.

So, what did we do? We started a GPON OLT FPGA design based on our own resources. Not something that we wanted to do, but the OLT takes a lot longer than the ONT to make. We knew we needed to start straight away if we wanted to have a product in the time frame that we wanted. I recall during the rest of my time at Tellabs that Broadlight asked us to design that FPGA out and our answer was no. Changing those designs is a lot of work and we had a functional answer.

Which is my business point here. We responded with what our partner told us with an alternate course of action. It was required for our business. No hard feelings. But the socket was lost right there, and there was never another entry point for winning back the business. So, if you customer gives you a direct request - you need to think long and hard when you give them a no.

Jim Sackman
FocalPoint Business Coaching
We Focus On Your Business - Time, Team, Money, Exit
Business Coaching, Sales Training, Web Marketing, Behavioral Assessments, Financial Analysis
Lunch and Learn Tickets!