In order to understand Internet really is we must see it in the context of the powerful idea it represents -- what happens when you’re forced to create solutions that minimize the dependencies upon the benevolence of others – in this particular case, service providers. The tragedy is that we find ourselves in the midst of an impassable regulatory thicket that seeks to assure we are dependent upon privileged service providers.
It’s a challenge to the very presumption that there is something to regulate. It’s not a question of whether we regulate or deregulation. We need to challenge dysfunctional entanglements themselves that define the tele/communications industry. The Internet demonstrates that these entanglements are not only unnecessary but frustrate our ability to solve problems and create value.
We need to step back and recognize that the Internet is not a thing and that the accidental properties of today’s Internet are no more than that – just accidental choices one makes in building a prototype. The benefits we associate with the Internet are really a consequence of a simple constraint – our inability to depend on the particulars of the path (AKA, the end-to-end argument). It is this philosophical shift that has liberated us from being bubble babies dependent upon service providers and forced us to create solutions which are far more powerful. One can call this an economic model but since I’m not an economist I’ll just call it a systems model
The upcoming FTC Broadband Workshop is on my mind as I read postings and news stories.
For example one story about the shift in so-called VoIP and another story on the so-called tiered Internet vs neutrality. Words like VoIP and Neutrality (and, for that matter, “Internet”) give the illusion of understanding without being specific enough to force the contradictions into the open. They are reminders that we can’t have a productive conversation if we use words that no longer have any shared meaning and if the words themselves betray a deep failure to understand the concept of the Internet.
The “Opportunity Dynamic” is simple but counter-intuitive. If you can’t pay for special solutions your only option is to take advantage of opportunities that already exist at the current or lower costs. This seems like a doomed approach since you can’t even pay to make a voice call – but that didn’t matter as long as you found some value – even if it was something as lowly as email.
In his Information Week article Mark Goldberg said that the Internet flourished because it was liberated from government control – that’s backwards. It flourished because the Web gave a lot of people a reason to care and to connect despite the phone network by using modems. When the paths were officially restricted the rules were difficult to enforce even for those who wanted to respect them and impossible for those who wanted to route around them. When I first connected corporate email to Internet/UUCP in the 1980’s I didn’t (couldn’t) make a big distinction between the two and mail seemed to just get though. I didn’t and really couldn’t know what paths the bits took.
If you need high quality bits to do voice then you’re stuck if you can’t pay for a special solution. The traditional response is to declare that End-to-End has reached its limit and it’s time to implement special hacks like MPLS to bring back the power of circuits or to implement 3G/IMS to lay out a path as per SS7. Even today the www.mocalliance.org tells us that the Internet was designed for data bits and that video needs RG-6. In fact, if you look around you’ll see a plethora of special transports like the recent 700 MHz public safety proposal to M2Z to the New York’s own $140m public safety radio debacle.
This is the importance of the discipline enforced by E2E (which I now call Edge-to-Edge to avoid confusion). It’s akin to the necessity of traumas to stress the immune system so that it could learn. We know the term “bubble baby” from the results of protecting the immune system from stress.
Fortunately we already had the PSTN which, though expensive by today’s standards seemed to be worth the price – thus there was no pressure to “fix” the Internet and allowed the marketplace dynamic to give us the web without a “web-surcharge”.
VoIP “just happened” – you couldn’t use QoS though you could try to assure that particular paths had sufficient fungible capacity. We had already managed to use the high capacity broadcast paths (cable and DSL) as IP transports despite their asymmetry.
Unfortunately we’re losing the PR battle by our willingness to accept the idea of what I call “Internet Inc” and raising the specter of a broadband gap. As a result those who are most dependent upon controlling the path and preventing connectivity are treated like heroes and are rewarded with the spoils. They can spend lavishly on private infrastructure and meddle in the networks within our home and do damage in the name of fixing the Internet so that it is locked into their purpose-driven service model. For me the awareness that Verizon is using IP makes me acutely aware of the shortfalls in the technology – as I try to navigate the set top box UI I wonder why the stupid (more like idiot savant) box can’t have a simple web server instead. It’s a very simple concept and should be trivial to implement but is symptomatic of the pervasive disconnect between the two worlds.
The tragedy is that even the net-heads seem willing to accept the idea that the Internet is a service and settle for locking it into fast pipes. Getting high speed connectivity is easy once you can simply pay for the fiber – in fact, we seem to be laying abundant fiber along every highway just in case because it’s so inexpensive to throw in a few more strands – to bad we don't do it for every byway too.
Instead we need to understand the power of the opportunity dynamic and work with it rather than frustrating it by accepting the idea of constricting ourselves within service offerings.
Part of the problem, perhaps, is the difficulty in understanding how these disparate systems composite by building upon success and sloughing off failures – a topic in its own right. The short explanation is that the opportunity dynamic is the inverse of whole-system engineering – since the only metric is that the system continues to function the failed elements simply don’t count because, by their nature, we don’t depend on them. Stated that way it sounds tautological but if you focus on the accumulation of successes then it’s simple and straightforward.
Efforts like COMMONS add to the commons/infrastructure model. We also need to improve edge-to-edge protocols that are not dependent upon the accidental properties of IP addresses (and its perverse cousin—the DNS). The millions of access points already provide coverage but await better protocols and applications that allow us to take advantage of the abundance.
My challenge is explaining this in a way that makes sense to those who still think only a phone company can create a call. I’ve purposely avoided getting caught up in the various disputes about the accidental properties of the CFR (Copper, Fiber and Radios). Unfortunately too many are caught up in the detail and can’t see that they are truly secondary to the dynamic. In a longer write-up I can cite many examples -- one I use is ISDN vs modems. In the end none of the details of ISDN really mattered – all that mattered was that it is treated as a valuable offering and modems and thus worth more. It demonstrated the difficult of raising the price in a real marketplace. Unfortunately broadband is ISDN without the alternative.
With this background, we should wonder why the FCC and the FTC feel it appropriate to assure we are dependent upon the privileged broadband service providers who can then divvy up the customers among them. Instead we should demand ownership of the basic raw material – Our CFR (Copper, Fiber and Radios).
