[BBLISA] last mile technologies

[Tom Metro]

Rich Braun wrote:
> Honestly, at this point the epitome of price/performance is a race between
> DOCSIS-3 and FIOS.

I thought FIOS had a superior network infrastructure, dispensing with
the neighborhood concentrators, which can be choke points.

Though it seems that the neighborhood concentrators are only problematic
if oversubscribe, and any network infrastructure is going to have
similar choke points, just further upstream, which, if not adequately
sized, will imposed similar problems.

On the other hand, the potential bandwidth from FIOS is way better than
DOCSIS, but if there is no pressure on Verizon to upgrade the
transceivers on each end of the fiber, that makes little practical
difference.

When you look at the bandwidth graphs published by content providers
like Netflix, the ISP's policies and willingness to properly upgrade its
backbone infrastructure seems to have a greater impact on actual
realized bandwidth than the last-mile technology used.

Several ISPs, particularly those in or soon to be in or near regions
where Google Fiber operates, have announced they'll be offering 1 Gbps
service to match Google.


> But I think a whole lot more construction is taking place
> with DOCSIS-3 than FIOS, probably because there are several companies using it
> whereas only one company supports FIOS.

Well, FIOS is a Verizon trademark. Google Fiber is fiber to the home,
and it uses similar, if perhaps incompatible tech. (I've never
researched who Google uses for their ONTs.)

Doesn't AT&T's "U-verse" product also offer fiber to the home? (I heard
recently that lately AT&T has been applying the U-verse branding to
their traditional DSL service in areas where fiber service isn't available.)

An then there is...

Brian O'Neill wrote:
> They even sold off some markets, like Northern New England went to 
> Fairpoint.

Correct. A friend is in NH and has a FIOS branded service from Fairpoint.

But I think the real reason why FIOS is not growing is...

Steve Meuse wrote:
> I believe I read that FIOS is not expanding to any new markets...

Correct. Verizon determined that margins were much better with their
wireless business (easy to charge lots per bit, impose caps, and if you
already own the spectrum, not huge capital costs), so they made a deal
with Comcast where they would sell Comcast TV service in markets not
served by FIOS, and Comcast would sell Verizon wireless in those same
markets.

Welcome to US broadband competition.


Rich Braun wrote:
> DOCSIS-3 has the advantage of lower cost:  it's exceedingly expensive
> to run fiber to every end-point, whereas running RG-6 from a
> pole-mounted fiber box to several nearby buildings is quite
> inexpensive.

I'd be surprised if the hardware for FIOS was really all that much more
expensive. It seems like DOCSIS gets its biggest win by repurposing the
existing coax that's already in place for TV.

The concentrator distribution model should provide some savings too,
though the cost of running a trunk line vs. a fiber bundle may not be
all that much, if you are starting from scratch.

I'd be curious to see what the FIOS vs. DOCSIS cost differences are for
wiring up a town that has no existing cable plant.

We know Google chose a fiber to the home approach, and they assure us
they have figured out a profitable business model for doing this, and
are not just offering this as a loss leader or market disrupter.

On any case, I think DOCSIS is winning because it is a cheaper upgrade
option if you already own a coax cable plant, and broadband companies
rarely compete in the same region, so no one is faced with building a
cable plant from scratch.

 -Tom

-- 
Tom Metro
The Perl Shop, Newton, MA, USA
"Predictable On-demand Perl Consulting."
http://www.theperlshop.com/