Fiber broadband, and its alphabet soup of flavors -- fiber-to-the-premise (FTTP), -to-the-node (FTTN), to-the-whatever (FTTx) --always catches my attention. It’s interesting, it’s the next big thing, and it’s terribly misunderstood.

First, there is this debate about AT&T’s FTTN under Project Lightspeed versus Verizon’s FTTP with FiOS. Who has the bigger pipe? Who can deliver more video channels? How much capex? Whose opex is lower? Overbuild versus greenfield?

Let’s check the facts, or, at least, the press releases. AT&T says it will invest about $6 billion to pass 19 million homes by the end of 2008. AT&T has since upped its goal to pass 30 million homes by year-end 2010, and it will, presumably spend a few billion more getting there. Customers can order the U-verse services suite that includes digital voice, about 10 Mb/s Internet access, and IPTV. AT&T says that it is marketing U-verse to about 50% of the homes passed. AT&T has always contended that adopting FTTN was the best time-to-market and capital investment model, given that most of its U-verse installations are in existing neighborhoods.

Verizon plans to pass 18 million homes with FiOS by the end of 2009 for an investment of about $20 billion. FiOS service includes POTS and digital voice, up to 50 Mb/s Internet access and up to 270 broadcast TV channels. Verizon is aiming for 25% take-rate for FiOS TV and 30% takes of FiOS Internet. From the beginning, Verizon believed that FTTP was the way to go for both overlays and new builds in order to provide big bandwidth pipes suitable for customer needs today and into the future.