[ NNSquad ] Re: [ PFIR ] New Short Video: "Network Neutrality in 30 Seconds" (Part 1)

[Barry Gold <bgold@matrix-consultants.com>]

Lauren Weinstein wrote:
> ------- Forwarded Message
>
> To: pfir-list@vortex.com
> Date: Sun, 24 Aug 2008 20:50:57 -0700
> From: pfir@pfir.org
> Subject: [ PFIR ] New Short Video: "Network Neutrality in 30 Seconds" (Part 1)
>
>
> 	 New Short Video: "Network Neutrality in 30 Seconds" (Part 1)


The comments below deal with the issues raised in Lauren's excellent 
video.  But I should also note that this video really doesn't talk about 
the core NN issue: packets is packets, and all packets are created 
equal.  Instead, it's about a related and important issue:  (possible) 
anti-competitive practices by cableco/ISPs.

Now... about the video...

This really does present the issues in a very brief format.  But
watching this brought home to me just what is too often ignored in the
NN debates.  So let's take it step-by-step using the same metaphor.

Cable/ISPs have a "hose" that can deliver X amount of water (bits) per
second.  They typically allocate 95% or more of this bandwidth to their
own video product(1), and the remainder to providing IP service(2).  If
a customer uses more than his "share", they will take some action(3).
But if a customer "uses" extra bandwidth via, e.g., PPV, or ordering a
higher-priced tier, that is folded into the charge for that service. So
in that sense, bandwidth caps are arbitrary and anti-competitive.

At the same time, they are following one of the standard paradigms for a
good user experience: waste what is plentiful, conserve what is rare or
expensive.  On a cable network, downstream bandwidth from the cableco's
"central office" (distribution point) to the customer is cheap.  They've
got _lots_ and _lots_ of gigabits available on the fibers (or copper
wires) coming out of their CO.  And that's all "fixed cost" -- it costs
them the same whether anybody uses it or not.

But Internet service is different, in two ways.

First, the ISP has to pay somebody else (a backbone provider) for a
"hose" (water main) to transmit their packets to the rest of the
internet.  If their customers demand more bandwidth, they have to buy a
bigger water main.  And that needs to be paid for somehow: either by
raising rates for all their customers, by charging the heaviest users
extra, by imposing bandwidth "caps" to encourage the heaviest users to
buy a higher service tier, or some combination.

Second, the current system is asymmetric(4).  The installed fiber/wire
base can carry multi-terabits downstream (CO to customer) but only a few
gigabits upstream (customer to CO).  This usually applies to the "hoses"
they buy from backbone providers too: a gigabyte of outgoing packets
costs significantly more than a gigabyte of incoming packets.

So it's perfectly reasonable for an ISP to seek to cap or discourage use
of the resource that costs them extra, while discounting the resource
(CO-to-customer bits) that is "already paid for".  Even without any
anti-competitive motive.

Another thing to consider: the economic benefit to the cableco.  Studies
quoted on this list have shown that cable service by itself makes very
little money.  It either breaks even or makes a tiny Return on
Investment.  That even includes the revenue for PPV services.

The real money comes from ISP-type services and cable telephones.  Those
produce an ROI that can attract investors, allowing the cableco to
invest in more capacity, extending their reach to more households,
creating new services to provide, etc.

So why should a cableco/ISP disadvantage the service that makes them
money (ISP/telephone) for the benefit of one that breaks even or makes a
minimal profit (TV cable)?  That would be insane.

That's not to say that large corporations don't sometimes act in insane
ways.  It's not uncommon to see a company busily pursuing a hopeless
business model while ignoring the money-making possibilities of
alternative/new business models.  Just look at the major record
companies, suing their customers in an attempt to defend a doomed
business model (selling songs on CDs through retail stores) instead of
figuring out a way to make money selling songs through the internet.

But the universe has a way of dealing with that type of insanity: sooner
or later one of two things will happen:
  a) the record companies will go out of business, and somebody else
will figure out how to make money in the new environment, or (more likely)
  b) one record company will figure out how to do it.  Then the rest
will scramble to copy the successful one -- and for a few years the
company that first really makes it work will dominate the industry.
(Until technological change comes along and requires yet _another_
change in business model.)

And that last is going to be the reality in _any_ form of information,
whether books, music, games, or anything else that can be represented as
bits: if you can find a business model that works for as long as 10
years you will be _very_ lucky.  Probably most companies will need to
change business models every 2-3 years.

(1) Channels that are included in the monthly fee, channels that cost
extra per month, and pay-per-view, combined.

(2) "ISP"-type services, cable telephone, and "two-way" services, combined.

(3) reduce the priority of their packets or charge them for excess use.

(4) true of current DOCSIS 2 installations.  DOCSIS 3 is symmetric, but
don't expect wide penetration for a few years.  Buying and installing
new modems costs money, so I would expect those to show up
   (a) in newly built-out areas, where there is no existing DOCSIS 2
modem to replace
   (b) for customers who are willing to pay extra for more bandwidth
   (c) as part of the normal cycle of replacing modems when they fail.


  [ One foundational problem is that since many of the associated
    bandwidth allocation factors and much key provisioning data are
    not routinely available for public inspection, and often are
    held by the ISPs as proprietary, it is difficult for most public
    observers to have enough information for straightforward
    determinations of possible anticompetitive behaviors by ISPs
    that aren't so gross as to be immediately obvious. 

      -- Lauren Weinstein
         NNSquad Moderator ]