The problem is that if you break down the costs, you'll find out that it almost doesn't matter what you put in as a cost of the total build; the big costs are the engineering and the labor to install, not the "cost of the NID" or anything like that. Nobody cares whether you saved a million bucks on a 2 billion dollar project. One of the cool things about the infrastructure that is now in place (copper pairs) is that it turned out to be relatively future-proof - lots of 50 and 70 year old OSP still in use. In order to get similarly long life out of newly installed fiber assets, the only real solution is home runs to either existing or newly constructed concentration points (not just a box at the side of the road, that's not what I'm talking about here). Distributed splitter designs force forklift upgrades when the Next Big Thing comes along, rather than upgrading the service only for folks who are willing to pay for it. The Next Big Thing is always coming, and 2.4 Gbit/sec down per port GPON is gonna look awfully slow 10 years hence when everyone's demanding gigabit ethernet to the desktop, not to mention 20 years from now with IPv6 multicast of 2000 channels of 4320p pr0n. I used to believe in the FTTC (fiber to the curb) model too - it's the "obvious" solution. That was before I started cranking the numbers myself, playing with some of the new splicing solutions that are out there that require *far* less finesse than the cam-splice stuff I was using 10 years ago. Now I believe in the "other" FTTC (fiber to the couch). Get it as far out into the field as you possibly can, right up to, or even inside, the house. -r Jack Bates <jbates@brightok.net> writes:
heh. I've seen 3 different plans for FTTH in 3 different telco's; different engineering firms. All 3 had active devices in the OSP. Apparently they couldn't justify putting more fiber in all the way back to the office.
Don't get me wrong. I've heard wonderful drawn out arguments concerning vendors that failed to properly handle Oklahoma summers or draw too much power.
Brings up new PRO: active devices in the OSP providing longhaul redundancy on fiber rings
Another PRO: simple, inexpensive NID
Jack
Robert Enger - NANOG wrote:
CON: active devices in the OSP. On 8/26/2009 12:06 PM, Jack Bates wrote:
jim deleskie wrote:
I agree we should all be telling the FCC that broadband is fiber to the home. If we spend all kinds of $$ to build a 1.5M/s connection to homes, it's outdated before we even finish.
I disagree. I much prefer fiber to the curb with copper to the home. Of course, I haven't had a need for 100mb/s to the house which I can do on copper, much less need for gigabit.
Pro's for copper from curb:
1) power over copper for POTS 2) Majority of cuts occur on customer drops and copper is more resilient to splicing by any monkey.
Jack