Re: Broadband == "Information Service"

(Some interesting analysis, plucked out of context, related to the 
recent 9th Circuit decision that broadband is not only an "information 
service."  -- Seth)

-----Original Message-----
From: Fred Goldstein <[email protected]>
Date:         Tue, 7 Oct 2003 09:33:13 -0400
Subject: Re: [CYBERTEL] questions about Legacy of the FCC's Computer 
inquiries by Cannon

At 10/6/03 11:42 AM -0700, Genny Pershing wrote:
>You might try looking at papers on the layered model
>of regulation.  Cannon's paper suggests that the
>Computer Inquiries follow this model.  The physical
>transport layer is regulated one way (as a non
>competitive monopoly market).  The virtual logical
>layer (the Internet) is regulated another way
>(unregulated title I).

This raises an interesting issue.  First off, I'll suggest that CI2 was
a
brilliant job, one of the most important decisions the FCC ever made,
and
it was largely responsible for making a public Internet possible, among
many other things.  But we have to put it in context of what was known
at
the time, and how the world has evolved, as well as knowledge of the
issues.

CI2 recognized the clear distinction, at least in 1982's market,
between
lower-layer transport activities and upper-layer information
processing.  At the time, there was a "natural monopoly" in most
transport,
with some competition for domestic routes between major cities, but
strictly enforced (natural or not) monopolies on most intrastate
routes.  There was no natural monopoly on higher layers.  So the rule
was
crafted to recognize where the telco networks ended, and to prevent
them
from leveraging that monopoly into higher layers.  It was very strong: 
In
the 1980s, while it was in full bloom, I was a member of the T1D1 and
T1S1
subcommittees, which wrote telecom (I worked on ISDN, Frame, and ATM)
standards, and our ITU counterparts (which we fed into) were writing
standards for "teleservices".  Those were carrier-provided services
which
ran above the bearer services.  The T1 side completely avoided them,
because teleservices were strictly off limits to RBOCs.  Oh, the good
old
days....

So there was a "Chinese Wall" at a layer boundary, and a pretty clear
understanding of who did what.  The layered model of communications,
which
was reinforced by the then-active OSI project, corresponded with both
an
industry structure and a natural boundary between monopoly and
competitive
services.

>When everything becomes an
>"information service" the distinction between the
>physical layer and the logical layer explodes.  Well
>those who are proponents of the layered model
>recognize that there is a difference between the
>physical layer and the logical layer.  And yes, when
>you fuse the two layers, they morph in to the
>competitive characteristics of the least competitive
>layer, which can leverage its position in the market
>place.

Precisely the problem... Regulations like CI2 recognized a distinction,
but
actually attacked the secondary characteristic (layer) rather than the
primary one (monopoly power).  And that persists today.  So we worry
about
whether a company is a "telecommunications carrier" vs. an "information
service provider", but ignore the far more meaningful question of
monopoly
or bottleneck power.  We used to control that indirectly, but the
monopolists have wiggled out of that jacket, and can potentially claim
that
their monopoly-power networks are "information" and thus not subject to
the
old rules.  This is totally bass-ackwards.

Regulation by layer or layer-based taxonomy has multiple weird and
ultimately unsupportable effects.  We have telephone companies, owners
of
monopoly local loops, claiming to be "information" providers (ISPs) and
thus using that to avoid their common carriage obligations.  We also
have
ISPs who meet a technical definition of common carrier in order to
lower
their costs.  In the latter case, of course, it was ISPs who set up
captive
CLECs (totally legal, of course) in order to change their dial-in lines
from subscribers to co-carriers.  When they started getting reciprocal
compensation for it, the ILECs had a hissy fit and the regulators (FCC
and
some states) stepped in to change the rules to protect the ILECs.

And the mother of all regulation-by-layer disputes is likely to be over
VoIP. In the Computer Inquiries context, IP was a higher-layer
information
service, but VoIP does quack a lot.   We have the phenomenon of
telephone
service providers throwing in what amounts to gratuitous sprinklings of
IP
in order to change their regulatory classification, or engage in tariff
arbitrage.  That isn't sustainable either.

And as noted, ILECs may be doing it in order to avoid the regulation
that
is appropriate based on their monopolies.  Rather than admit to being
monopoly telephone companies, they will claim to be non-monopoly ISPs,
but
of course that's just a dodge.  The rules should be based on market
power,
not on layer.

BTW on pure technical grounds, the OSI model is obsolete, and TCP/IP,
which
corresponds to it, is too.  Basing regulation on a 25-year-old lab hack
protocol suite is also very dangerous, even if that suite has, in
effect,
monopoly power of its own.