Re: H.R. 3458, Rep. Markey's third bill to stop telecom from mismanaging our basic infrastructure.

[David Farber <dave () farber net>]

Begin forwarded message:

From: "Bob Frankston" <bob2-39 () bobf frankston com>
Date: August 25, 2009 10:51:23 PM EDT
To: <dave () farber net>, "'ip'" <ip () v2 listbox com>
Cc: "'Lauren Weinstein'" <lauren () vortex com>
Subject: RE: [IP] Re: H.R. 3458, Rep. Markey's third bill to stop  
telecom from mismanaging our basic infrastructure.

I am not going to convince Brett but he is voice objections that  
others have so I’ll take advantage of this opportunity to respond to  
the larger audience. I’ve covered much of this in http://frankston.com/?n=BI 
 (or http://rmf.vc?n=BI for the URL-challenged).

**One cannot win these arguments by pure reason – you need a  
demonstration. The Internet has done just that far better than we  
could have imaged. Yet as long as we give the existing telecom  
providers control over the transport these myths will fester. Worse,  
given that their very existence is premised on being able to add value  
to the network they must not accept the idea of treating bits  
indifferently.**

This is why it is so very important to separate the transport from the  
applications. You could argue that the original separation of TCP form  
IP was necessitated by the inability to guarantee that the transport  
could serve the needs of the applications. The end-to-endargument said  
that you can indeed create solutions at the edge of the network. You  
wouldn’t have been able to convince people to give up the idea of  
circuits without the Internet being such a powerful demonstration  
though Louis Pouzin did try.

(As I wrote in http://frankston.com/?name=InternetDynamic) the end-to- 
end constraint – the inability to depend on the network – was a  
powerful driver that got us past all the “obvious” assumption that you  
could only solved these problems if you build the features deeply into  
the network. At first you couldn’t reliably do voice. But we didn’t  
have to – there was already a voice network. Over time all it took was  
more capacity and voice just started work and it worked at a zero  
marginal cost because you couldn’t have a separate voice network. Same  
for video.

**This is why it is so very important to enforce neutrality – to avoid  
falling into the same traps. As long as we build in special hacks it’s  
hard to prove that the are unnecessary.**

We need to ask why people are so darned determined to build in the  
legacy assumptions. If we already have fine video distribution network  
already why must we reshape the Internet to act like a video  
distribution network. Just use the one we have. The fear seems to be  
that the Internet works so well that we can’t economically maintain a  
single-purpose network that is no better than one that does it all at  
essentially no cost.

**As to saying that customers want to buy all their services from the  
one place…**

If you don’t want to drive, then hire a driver or take a bus. But that  
is no reason to say that all transport must be owned by a railroad and  
that we must all buy tickets.

**As bandwidth becomes more of a commodity there is indeed no way of  
differentiating themselves.**

Yes, the current business model would fail in a real marketplace. This  
is the point the carriers themselves make inhttp://frankston.com/?Name=AssuringScarcity 
. So why do we try so hard to keep it going no matter what the damage?

This is why the very idea the we must use networks as if they are  
railroads (http://frankston.com/?n=RailRoads) doesn’t make sense  
anymore. We should view networking more like highways or a bit  
commons. This way we would all be able to create solutions and we’d  
get the benefits of the maximal network effect.

**The huge deficit a the reason we must stop funding the telecom  
industry by forcing people to pay hundreds of billions for services  
they can do better on their own. And we lose much money by forcing all  
new ideas to justify themselves to a gatekeepers artificially high  
prices.**

I’ve addressed the cost issues in http://frankston.com/?name=IPTelecomCosts 
. If bits are bits then we do we pay a premium for voice bits? I’ve  
compared this business model to forcing people to pay a premium to  
take the 59th St Bridge vs. the Queensboro in NYC. Once people realize  
they are the same bridge they would be outraged. Our policies are a  
multibillion dollar subsidy to a synthetic industry with incalculable  
collateral damage in our lives and our safety.

We’d get hundreds of billions of dollars, if not a trillion dollars,  
into the economy if we stop trying to fund a telecom industry and  
instead use “IP” (and other) connectivity which is just the way we use  
existing resources. We already have huge untapped capacity in our  
existing wire and fiber and our wireless paths. So let’s stop these  
stale and dead arguments about how much it would cost to subsidize the  
Internet. That’s the argument made in the 90’s when people though  
phone calls had to cost 10¢ a minute to reach the next town and though  
the Internet must be a trick.

And let’s get past the utter nonsense of a “government owned  
infrastructure”. That is what we have now – government giving  
privateers a charter for exclusive control of the public rights of  
way. The alternative is nonexclusive use of community resources. Even  
more so for wireless (http://frankston.com/?n=SpectrumDirt) which  
would no longer be constrained by having the government locking it  
down so it would be resold the private owners who, in turn, can charge  
us for using what costs them nothing other than the license fee and  
the gear they control. (http://frankston.com/?name=IPPotHoles).

**The idea that we shouldn’t guarantee particular services is one of  
the more subtle issues.**

This one of the reasons for antitrust – we need to create the  
opportunity for new ideas. We don’t know what ideas will be important.  
What if we had to build the web into the phone network? The reason  
things are confusing is that the Internet does indeed subsume telecom  
so we then get nervous and now want to bake in telecom emulation. Yet  
doing that is like baking the golden goose – no goose and no eggs. The  
Internet is a dynamic and that is very hard for people to comprehend.  
We’ve had only one or so generations of  computing – the language and  
understanding of dynamic systems is still new. It’s more like the days  
of Audubon who had to kill the birds he loved because it was the only  
way to study them.

**But I’ll never convince Brett and others who think of the Internet  
as nothing more than another telephone company.**

Again, this is why we have to separate the content from the  
infrastructure. The Markey Bill may not be perfect but it is far  
better than continuing the allow basic misunderstandings keep us mired  
in the past. We must let specious reasoning keep us mired in the past.

PS: Where is Jon Stewart when I need him 
http://www.huffingtonpost.com/2009/08/21/betsy-mccaugheys-ideas-ca_n_264970.html?

-----Original Message-----
From: David Farber [mailto:dave () farber net]
Sent: Tuesday, August 25, 2009 18:31
To: ip
Subject: [IP] Re: H.R. 3458, Rep. Markey's third bill proposing to  
heavily regulate and micromanage the Internet



Begin forwarded message:

From: Brett Glass <brett () lariat net>
Date: August 25, 2009 5:59:34 PM EDT
To: dave () farber net, "ip" <ip () v2 listbox com>, bob2-39 () bobf frankston com
Subject: Re: [IP] Re:   H.R. 3458, Rep. Markey's third bill proposing
to heavily regulate and micromanage the Internet

At 02:09 PM 8/25/2009, Bob Frankston wrote:

> I see this bill as a step towards deregulation  telecom regulations
> tend to be obsessed with slicing and dicing the kinds of services and
> pricing. Without being burdened with dealing with each service there
> is very little regulation  just the assumption that we solve problem
> by providing more capacity and that's it.

Alas, this is a false assumption. Just as it's foolhardy to assume
that the computer with the most "megahertz" is the fastest, it's
dangerous to assume that adding capacity will solve all problems. All
the extra bandwidth in the world, for example, will not make up for
long latencies, too much jitter, or too many dropped packets. Service
providers must be able to sell quality of service, just as FedEx and
UPS can sell expedited delivery or a trucking company can sell
refrigerated transport to prevent food from spoiling. One of the most
serious problems with the Markey bill is that it actually prohibits
this.

> The easiest way to meet the provisions of the bill is to have the
> transport business completely separate from the content business.

In the real business world, this is not possible because consumers
desire and expect "one stop shopping." They expect their ISPs to
provide them with electronic mailboxes, Web hosting, tech support
(often for their PCs, too, even though this falls entirely outside the
scope of an ISP's business!), and -- in the case of mobile broadband
-- mapping services, SMS, directory services, voice mail, and more.

What's more, as bandwidth becomes more and more of a commodity, these
ancillary services become the providers' only means of differentiating
themselves and/or innovating.

The Markey bill does have a clause that purports to prohibit tying of
Internet service to the purchase of other products. This is a token
provision, however, because no one has done this since the telephone
companies finally agreed to allow "naked DSL." And because the clause
fails to prohibit "bundling" (which both the cable companies and
telcos are doing; one can expect to pay a 50% premium if you buy cable
modem service without cable), there is nothing in it for the consumer.

> Whether the bill passes or not it would be very useful exercise to
> think about the implications of such a separation. Reduced to pure
> transport it would become obvious that we have redundancy rather than
> competition and we'd see rapid consolidation into a common transport

This sounds like a proposal to drive commercial ISPs out of business
via complete commoditization of their products, leaving behind
government owned infrastructure

> While I would prefer a simpler bill that explicitly said bits are  
bits
> and saying that we should fund infrastructure as infrastructure,

With the huge deficit and national debt (which have been such a large
issue in the debates over the economic stimulus and health care
reform), do we really want to fund Internet out of the public purse
when private parties are already willing and able to do it?

--Brett Glass





-------------------------------------------
Archives: https://www.listbox.com/member/archive/247/=now
RSS Feed: https://www.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com




-------------------------------------------
Archives: https://www.listbox.com/member/archive/247/=now
RSS Feed: https://www.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com