Services vs Infrastructure!

Read Opportunity for an overview. The whole concept of "telecommunications" seems to be fundamental and necessary. But if you step back you see how all the pieces fit together but they are in a world of their own. We are able to create our own solutions. The question is not whether carriers will permit us to communicate, it's only a question of how long will we allow our economy and safety to be held out bay in order to support an obsolete and business model that preserves scarcity rather than allowing us to get the benefits of abundant connectivity.

16-Jun-2006 Bob's other writings. Version 2: 2023-05-28 17:09:26

Too Simple to See?

I feel like the boy in the crowd pointing to the emperor and asking why he has no clothes. This time the clothes are magnificent but the emperor is an illusion.

The complexity of the FCC and its Regulatorium provides an impenetrable barrier of stultifying complexity that intimidates people who feel they must master the arcane complexity of its obscure rituals lest they seem unworthy to participate.

I assure you that I’ve had forty years of experience and there is no there there in the Regulatorium. Just a lot of details that keep you so occupied that it’s easy to ignore the fact that the defining principles are not longer true.

Underlying the concerns about “network neutrality” and “broadband” and “Internet Policy” is a one very simple point. The entire telecommunications industry is premised on the assumption that there is an intrinsic relationship between the transport of bits and their meaning.

We now know that this is not at all true and any attempt to use the control of the transport to restrict communications and restrain trade is an egregious violation of antitrust principles and causes grave economic damage and creates real dangers in preventing us from solving our own problems and responding to events.

Unfortunately the fiction that transport and meaning are the same is codified in what I call the Regulatorium – the system of rules that established the Federal Communications Commission and the industry it is charged with protecting at any cost. It is as if we were playing a game of second-life and confused it with reality.

The Regulatorium’s broadband policy is akin to funding more railroads. This is backwards. It’s our infrastructure like roads, sidewalks and even rivers. Telecommunications frames the problems in terms of services and value comes from scarcity.

By framing the problem as infrastructure we build the services ourselves and have the incentive to fund private, local and regional infrastructure to meet our own needs. Value comes from abundance.

We should have a straightforward positive agenda. Instead of having to ask for neutrality as a concession we must recognize that it’s our infrastructure and we will decide how to use it.

This should be obvious. The question is why most people would rather beg for concessions than take control of their own future.

Part of the reason is a lack of technical understanding – it’s easier to ask for more of the same and plead for concessions.

It’s all too easy to gets lost in the stupefying and arcane details of the Regulatorium. It’s necessary to step far enough back to see the whole. Only then does it become obvious that the whole regulatory system is detached from reality. The Regulatorium and its associated industries are like a cult – very real for those inside and meaningless from the outside.

The current History Channel series on the American Revolution is apropos. Ben Franklin didn’t start out as a revolutionary but he learned that there was no middle ground. We must choose between a marketplace and an oligopoly.

The ATT Long Lines building at 33 Thomas St is a 50 story windowless building erected in 1974. It is imposing but its function can be replaced a few inexpensive routers. This is the plight of the entire industry. The price/performance improvements in computing during that period exceed our imagination.

In 1974 a megabyte of memory cost a million dollars. Today a thousand times that much costs as little as $10 (as flash memory). That’s a cost reduction of a hundred million times. Every element of computing and communications has shown that kind improvement!

Why are we still paying many billions of dollars for an industry that hasn’t changed much since 1974? Additional services like broadband are nice but that 1974 copper wire is now capable of carrying millions of bits per second but the price of the copper has gone up for some reason.

VisiCalc taught me that numbers don’t lie, they just lay there. It’s all in the interpretation. I was reminded of that recently on a segment Penn & Teller’s show, Bullshit, as they demonstrated how we are easily led astray by numbers. A 600 page book full of charts and graphics is meaningless if the premises are false and we know that the premises of the Regulatorium are false. Don’t let them befuddle you with reams of well-fertilized “data’. If you demand charts and tables you can do no more than hold a mirror up to the past. Remember Troy and beware of charts containing agendas.

Transcending Telecom

Even as I complain about the current “telecommunications” industry I am an optimist. I paint a negative picture of the current industry to emphasize that we shouldn’t try to fix it. Instead we should focus on assuring that we have the opportunity to do far better.

My challenge is two-fold. First helping people see past the illusion of today’s tele/communications industry. Second, helping them understand that it’s about simple infrastructure that can be funded locally and regionally.

This should be obvious when we compare a moribund telecommunications with the rapid growth of the personal computer industry and the Internet itself.

If I seem impatient with the attempts to make incremental attempts to fix the current telecommunications industry it may be due to my experience both as a programmer and as a participant in the rapid changes in computing and communications.

As a programmer I know that when fixes create more problems than they solve we need to rethink the design. Legislation to reform telecom seems to create more problems than it solves.

As a student I studied the telephone network and watched the development of the technologies that gave rise to the Internet.

As an entrepreneur I saw the impact that VisiCalc had on the financial world in the 1980’s. It was only 32K bytes of code yet it was at just the right leverage point to have a major impact on finance and computing. Neither Dan Bricklin nor I anticipated what would happen but it did make us appreciate the impact a simple idea could have.

In the 1990’s while I was working at Microsoft I intentionally set in motion a marketplace dynamic to bring the Internet to all the computers inside our homes. The conventional wisdom was that the carrier’s triple play offering would define a future in which we’d all be watching TV with entertainment and services coming through our set top box. Simply removing the impediments to connectivity set the triple play back by a decade.

Making IP connectivity available meant that anyone could reach the customers directly without having to pay a rental fee to the carriers. Yet the carriers are still trying to demand the rent – that’s why there is such concern about neutrality. Far better not to give them the opportunity in the first place.

I’ve worked hard to understand how marketplaces work. In 1997 I wrote a chapter for the book, Beyond Calculation. I wrote about the limits of Moore’s law. I explained that it wasn’t at all about physics but about marketplaces. I call my version of this the Opportunity Dynamic or, perhaps Frankston’s Law – marketplaces that create opportunity rather than just giving us solutions (or services) allow demand to create supply.

When I argue that the problem is the fundamental structure of the telecommunications industry I am not making a casual observation nor bashing the carriers. I am applying principles that I have tried hard to understand. These are the same principles that have given us home networks and made networking gear a consumer item.

My technical background enables me to understand how connectivity works – why the Internet works despite, not because of governance. It is a dynamic that means we don’t need nor want carriers giving us the solutions that are profitable to them.

We should limit ourselves to asking for more favors like broadband. We then find ourselves having to bargain for neutrality rather than assuming it. Why aren’t we asking for ubiquitous wireless coverage rather getting excited by ring tones?

Why don’t we just run our own fibers, use the copper to its full potential and deploy our own wireless access points?

Why aren’t we seizing the opportunity instead of asking for services? You can’t ask for the web – you need the opportunity for a lot of people to try out ideas and maybe one out of six billion people might get it right!

We need to recognize that the Regulatorium is the embodiment of technical misunderstandings from the 1920’s and the depression era fear of capitalism. It must not be allowed to define our future by assuming that only they can create services and denying people like Tim Berners-Lee the opportunity nurture a small idea that seems to have no value until it grows up.

We must allow the FTC to enforce antitrust so we are able to fund infrastructure and create our own solutions transparently and honestly.

The rest is a detail but if you want the details …

Beyond Neutrality

Today we frame the issue in terms of neutrality but that assumes that we are not in control but are limited to a choice of services.

The issue is not neutrality. It is whether we will continue to fund our infrastructure by capturing the value of services and limiting availability in order to maintain prices. I compare this with buying drinks you don’t want in order to watch a floor show at a club.

There is an alternative – honest transparent funding of the infrastructure as such with each community contributing to the whole. This is a dynamic that encourages abundance and creates value as we create our own solutions.

Once we understand that we then realize that the Internet is not a network but an idea. It is the ability to create our own solutions using any available transport rather than being dependent upon network providers for our services. As I explain in “Telecom Phrase” the transport and the services are not decoupled. Businesses that use their control of the transport to force us to buy their services should be subject to antitrust scrutiny. They are open and explicit about their plan to assure scarcity.

Unfortunately the carriers are protected by the assumptions of the Regulatorium that lock in the na�ve assumptions of the 1920’s and 1930’s into statute. We can compare the situation to the S&L debacle of 1980 when the US Congress had to admit that it had set up a problematic dynamic and had to cover a trillion dollars in bad loans.

In the 1980’s we recognized that misguided policies had lead too many Savings and Loan Associations (essentially banks) to make imprudent loans in order to stay competitive. By the time we faced up to the crisis there were a trillion dollars in outstanding loans.

Today’s telecom crisis isn’t as obvious because the funding model seems to be working and we can’t measure the opportunity we’ve been denied. What is important is that we can facilitate the transition from the old world of service-defined telecommunications to one in which our infrastructure provides opportunity for us to contribute. The biggest difference is that there will be a net cost savings in such a transition because by treating the infrastructure as a common transport rather than having a separate infrastructure for each purpose we will greatly reduce our costs.

Paying the carriers for their current assets would be generous because they will lose their market value in the face of competition from our ability to use our infrastructure as a fixed asset. An offer to pay, essentially a bridge-loan for the economy would allow us to get the benefit of transparently funded infrastructure and the benefit of being able to create our own solutions. The time value is too great to justify waiting.

We can start to get immediate benefit by doing what we are already doing – using the Internet Protocols over our existing infrastructure but without the burdens and limitations of being forced to choose among carrier-provided services.

The situation is coming to a head as we become more adept at taking advantage of any available connectivity. The stark reality will soon become apparent to investors despite the optimism of the latter day Grubmans. It’s no surprise that analysts are bullish on telecommunications – analysts don’t normally choose to report on industries that they don’t believe in. Analysts like James Enck who question the future of the industry are the rare exception.

The industry looks formidable and necessary. It must be funded otherwise we’d find ourselves unable to communicate.

The industry has been hollowed out and, as we will see, all the costs are real but they only exist within their closed reality. The fiction cannot be maintained much longer and it’s all about to blow apart. The rage over net neutrality is a hint of the coming change as we learn that we don’t need nor want companies the products of the Regulatorium – the telephone and cable companies. As communities and municipalities discover that they can provide their own connectivity and interconnect with each other they also learn that they are in a position to gain far more than a carrier funded by service revenue can possible provide.

The carriers’ assets have no intrinsic value – they just own glass and copper and some radios. We create the value. The abundance of fiber capacity was costly to investors.

As we become more adept at making effective use of any transport available this dynamic is going to extend to the entire industry – both local and regional transport.

We should be generous and offer to finance a transition from a service-based model to infrastructure. It would be a wise investment and greatly reduce the costs.

The investors need to understand this dynamic. Otherwise they may be forced to confront reality when they find that they cannot maintain scarcity and they will find they own nothing.

From Analog to Digital

In the days of analog telephony the phones worked very hard to make the whole system work. Over the decades there were many innovations in operating the network and assuring that you could hear the other person over a long distance.

It was expensive to build an infrastructure that worked at all, let alone worked well. The idea of having a single company manage the entire network seemed to be a success and the US system was the envy of the world.

It’s no surprise that the phone companies played an important role in developing digital technologies culminating in the Intelligent Network or Signaling System 7 (SS7).

SS7 is designed according to classic telephony principles. The assumption is that resources are scarce so it’s worth putting a lot of effort into creating a path or circuit. If the resources are available a path is established and maintained. The system itself has two layers – the inner layer which provides the basic network and communications functions and an outer layer for the trusted computers that provide services such as voice mail.

SS7 was designed to handle only telephone calls. The routers than can replace the 50 story building at 33 Thomas St can do much more. As David Isenberg observed in The Rise of the Stupid Network, those routers are not intelligent like SS7. Therefore they don’t know what the bits they are carrying mean and thus they carry any kind of traffic or content.

It’s difficult to tell investors that all the switching offices are now liabilities that need to be decommissioned. It’s better to do what normal business do – focus on the future and hope that people won’t notice the present.

The old intelligent network is recast in a new role at the center of convergence.

Convergence!

The same Internet Protocol that allows the gigabit switch to carry all kinds of traffic allows the phone companies to offer new multimedia services.

Cellular telephony has been the proving ground for this idea because the cellular carriers were forced to adopt advanced technologies in order to operate a large dynamic network. Digital packets technologies were the basis for the huge expansion in the use of cell phones and the same technology allowed the carriers to offer a whole variety of new services. It took the carriers years to make the network operate so you could roam from city to city without having to dial special numbers.

If you look at the protocols used for the cellular network they look familiar. As with SS7, 3G maintains relationships between phones though it has to handle mobile devices and manage resources dynamically. The protocols for messages are based on standard IETF protocols. But they are improved because the carriers have control over their network and know who the users are.

The same technology is being brought to landline telephony as IP Multimedia services.

With all the services converged on the carriers’ network services they are able to stay competitive and make money selling these services or IMS.

Quality of Service (QoS) has been a defining concept throughout the history of telephony. At first getting any quality at all was a challenge. Over the years QoS was quantified to assure the best experience while using preserving scarce resources. Even as network resources have become plentiful the concept of QoS has persisted has an unquestioned assumption. Even if voice traffic becomes less of a problem we need QoS for streaming video.

If QoS is important then one can use it to define tiers of service.

Cable Companies had a different history. They started by bringing broadcast TV signals to those who couldn’t get good reception by placing the antennas on hills and containing the signals within a coaxial cable. Each signal had its own band, an exact image of the broadcast signal. This is the origin of the term “broadband”.

The cable TV systems have evolved. Once you have your own coax you aren’t limited to the broadcast channels and can offer a wider variety of content.

The phone companies saw this and developed ADSL so they too could offer high value TV services but the technology wasn’t up to the task. Today they are trying again with faster DSL and fiber (as in Verizon’s FiOS).

With the advent of the Internet they both took advantage of digital technologies and offered Internet connections as part of their product mix.

Managing a complex phone network is expensive. Paying for all the video content is expensive. Most expensive is the need to pay for all of the infrastructure they have to deploy to keep up with the demand for their services.

The cellular companies continue set the pace with new services offerings on the cell phone which remains the symbol of convergence. The requirements of spectrum management make the carriers responsible for assuring everything works just right. With responsibility comes control. Only the carriers are trusted with operating the phones and thus they are the only ones who can define new services.

In looking at the carrier landscape we mustn’t forget broadcast TV. The introduction of spectrum auctions for cellular telephone demonstrated that spectrum is a valuable resource. While the rest of the industry went digital television stayed pretty much the same as it was fifty years ago.

It makes very inefficient use of a huge portion of the spectrum. In the 1930’s television was a great engineering feat and only worked because every element of the system worked just right. Just as it took an act of congress to create the television broadcast system (via the FCC), Congress is playing a central role in upgrading television to digital and using the opportunity to free up valuable spectrum. Digital technology creates new problems – it’s too easy to copy the content so Congress “solves” the problem by introducing a “broadcast flag” which prevents copying. Well, it doesn’t prevent copying, just makes copying or modifying the bits a federal offense.

Digital TV is DTV. High Definition TV is HDTV. This can be confusing and is confusing. HDTV would simply be a new product like Quadraphonic sound. It’s a 1974 technology but it’s difficult to deploy until the entire broadcast system is upgraded. No one understands DTV but HDTV seems compelling so it’s easier to tout the importance of HDTV as if it were an essential industrial policy.

We have a complex set of services and an expensive infrastructure funded by selling the services. This includes “Internet” as a service.

In Assuring Scarcity I point to the cellular industry’s presentation in which they emphasize the importance of assuring that people continue to buy their services.

This is vital for the viability of the industry. If they can’t differentiate themselves and everyone had the same access to the bits that the carriers do they wouldn’t be able to fund the infrastructure.

Carriers are used to spending billions of dollars on infrastructure but there are no longer any guarantees as companies like Global Crossing and Cable and Wireless have discovered.

Even though the carriers may be making all the investments they promised, they continue to spend billions on new infrastructure such as broadband under the assumption that they will continue to be able to earn service revenues. This has worked so well that it’s easy for them to feel they deserve it. For without that income who would pay for the expensive infrastructure. Who would operate the network?

They need to be able to take advantage of their control of the transport in order to get a portion of the value created using their transport. They can’t be neutral.

No wonder they are trying so hard to prevent neutrality – even if it’s hard to define.

Divergence

But they can’t prevent the value from leaving their network because the importance of the Internet is in enabling divergence – the creation of all sorts of new solutions and applications rather than an enumerated list of converged services.

Neutrality can’t prevent people from finding opportunities and taking advantage of whatever is available.

The fundamental economics of the industry are at ends with itself. You can’t make money running a canal across an ocean.

I tried very hard to tell the carriers “convergence” story as sympathetically as possible. It was very hard because it’s all complete, consistent and nuts. That’s a harsh word but it only makes sense if you don’t step back and see that the defining premises are false as I explain in “Telecom Phrase”.

It’s easy for carriers to believe they have a necessary role. After all, people don't really want to write software and solve their own problems. This is the argument people made about personal computers – they were just too hard to use.

The devices were indeed hard to use and needed software to make them usable. But you couldn’t predict who would write those programs so you had to make sure everyone could. The few of us who did were able to provide them to others who could run it on their own machines.

The same is true for the raw Internet. Today you can use Skype or Vonage or other companies’ products and services to make phone calls. This is the Opportunity Dynamic in action.

While many people did write their own word processors the real value came from the unanticipated applications like the web.

The industry argues that it needs control over its networks in order to have an orderly marketplace in order to be profitable. In the presentation I point to in Assuring Scarcity, that if there was abundance then anyone could create their own solutions and thus no one will. Perhaps they misunderstand the philosopher Yogi Berra who said that no one goes to his club anymore because it’s too crowded.

It’s a nice parody of elitism but an offensive argument from companies that claim to be real businesses.

It’s all too easy to find negatives – reasons why new approaches can’t work. When the alternative is frightening there’s little incentive to ask hard questions. Yet the carriers knew that something like Voice over IP was coming but couldn’t deal with the concept – it was easier to tell regulators and others that only a carrier could provide such capabilities. And, to this day, they maintain the fiction in order to keep their users creating billable events.

After ATT was broken up people were allowed to own their own telephones yet many people continued to pay a rental fee. After two decades the courts severely chastised the carriers and fined them. At least there was a physical object being rented. Charging for phone calls that have no intrinsic cost is far more outrageous.

The Regulatorium is so complex that it absorbs all ones attention. If you have to live within the Regulatorium you must live by its rules as if you were playing a game with high stakes. Just learning the rules is a life’s work.

I saw a surreal example of this when attending back-to-back panels. The first panel discussed FCC rules and how they impact the purchase of switching gear. The very next panel, with some of the same participants was about voice over IP and the panel observed that no one was buying any of that gear.

Surreal, yes, but it is treated as reality by all those involved. It’s our economy and safety and our future yet the legal profession is required by the rules of their profession to act is if the Regulatorium is meaningful. I feel like Galileo trying to make his case before the medieval church.

I understand why the rules exist but I don’t understand the failure of leadership to rise above the details and see the forest for what it is – a barren island full of artifacts but devoid of life.

This centralized design allows the carriers to sell services and thus make the money to pay for the complex systems which is necessary in order to create the services to sell. All tautological and therefore fully self-justified even if none of it is required.

In computer science this is a frequent problem – you allocate elements of a system which reference each other. They all look necessary because of these references. We solve this problem with a technique called GC or Garbage Collection. We trace all the references and find all objects that are referenced by each other but not by anything outside the circle and discard them as garbage.

Time to GC telecom.

The Opportunity Dynamic

The problem for the carriers is that the Regulatorium cannot save them from what I call the Opportunity Dynamic or Frankston’s Law. It’s not just that they face hordes of innovators competing with them; it’s that there are many kinds of problems to be solved.

This is a very real process and has stark consequences for the telecommunication industry which is based on the ability to own the value created using its transport.

At a recent Massachusetts Technology Leadership Council Meeting (MassTLC) meeting Beth Comstock of NBC Universal spoke about their digital strategy which includes downloading complete story lines over the Internet and they are heading towards using it as a standard distribution channel with traditional TV being just an option. Downloading is not at all sensitive to the kind of tiering the carriers are trying to offer.

What she didn’t mention is that we don't need to turn everything into video files. Adobe’s Flash and future technologies are many times more bit-efficient than traditional video because all they need to do is send some descriptions and describe movement. But the Flash players are typically run on general purpose PCs. They are software platforms which aren’t subject to the carriers’ veto.

Even traditional services like classic phone services can be done better at the edge. I had a dramatic example recently when flying from Stockholm to Boston. I used the Internet connection on the plane to call home using Skype. The connection itself was limited but my wife didn’t even realize Skype was using buffering and other tricks to improve the quality. The lack of any cost above the fixed cost of the connection for the flight was in sharp contrast with the satellite phone nearby that would let me call for $10/minute with far less quality. (Notice the Hotspot sign and the phone across the aisle). In reading this my wife reminded me that she’s hard of hearing which makes Skype’s performance even more impressive.

The carriers are assuming that QoS (Quality of Service) is important and even necessary. That has been their obsession for a century and they’ve done a remarkable job. You can now call someone anywhere in the world and hear them clearly thanks to digital technology.

The problem is that it’s no longer a hard problem. It’s actually far worse as we see with the airplane example. On many international calls the quality of a VoIP call is often far better than traditional telephony which is hobbled by the historically high cost of such calls. The capacity is no longer limited but the carriers know that the costs have come down and they can’t afford to improve their own infrastructure. The best they can do use the same IP transport that the rest of us use.

The VoIP calls do not depend on QoS – I confirmed this with Tom Evslin who provided some of the initial commercial grade VoIP transports!

The carriers don’t control enough of their network to even define quality across a dynamic path. The best they can do is control the quality of the first mile connection between their customers and their central offices.

While we can’t define net neutrality let alone mandate it, the rage is real. It is difficult for the carriers to make dramatic changes. As much as I and others complain about the lack of real competition, there is sufficient bit-level competition to limit the carriers’ ability to play too many games with IP traffic.

Even as Excite@home accused those with Webcams of abusing their service, other carriers were touting the capability. Today carriers may block port 80 in hopes of preventing people from running web servers at home but third parties can route around this.

As we see with the example of NBC-Universal, streaming is not a big problem because most content can be downloaded even at slow speed and await the user on the PC. Downloading means there are no channels or television grid to limit the capacity!

While Congress looks forward to the bounty of the spectrum auctions all innovation is moving to the unlicensed local packet radios where there is no tax on innovation. We can then use the Internet to send a message any distance. There is no distinction between wired and wireless – bits just need a path. Because we only need to communicate locally the capacity of the sliver set aside for unlicensed uses like 802.11 has essentially unlimited capacity.

Over the Air (OTA) television is only the trailing 15% of the marketplace and not at all interesting anymore. Even more so as Internet connectivity becomes ubiquitous

Unfortunately the “Broadcast Flag” is the kind of interference that does real damage in the marketplace by making innovation a federal offense and forcing us to use obsolete technologies unless Tellywood deigns to give us permission.