You may be surprised to learn that it had nothing whatever to do with you and I, the consumers of telephone services. From the telco's perspective, you are, to a degree, a captive audience that they already keep under control by means of restricted signalling at the network edge (Part 2) and various non-technological means, including government lobbying. When many consumers chose to ditch their wired phones in favour of a mobile phone, the telcos bought the cellular phone companies, further restricting alternatives.
IN was instead driven by the consequences deriving from the love-hate relationship between the telcos and their suppliers, in particular the major manufacturers like our own Nortel Networks.
Like in anything having to do with business we need to follow the money to understand this point. Observe the diagram on the right. Here we see the consumer, you, as the source of all revenue. Since a little over a decade back, you can select from one of very few service providers. They in turn spend your money on the equipment and services needed to operate those services you've paid for. In an ideal world, you are satisfied with the value you are getting for your money and the telcos and their suppliers all make a reasonable profit.
Dig a little deeper and you find more forces at play than financial balance sheets. You depend on the telephone company, and you now have some, but not a lot, of choice. Similarly the telco has a dependence on the suppliers of the hardware and software that comprise its network and services. This latter dependence is at the heart of IN.
The switches from Nortel and Lucent, among others, are not simply switches. They also contain the software and the signalling endpoints that enable the bulk of the services you use. Because these switching platforms are closed systems the telco must request features from the manufacturer. The process is long and tedious as the parties negotiate, with the manufacturer looking for an up-front commitment on sales volume before even beginning development.
The telco has other difficulties. Since in North America they pretty much use the switching equipment from just two suppliers: Nortel and Lucent. Since they use both it does them little good if, say, Lucent agrees to build a new service when Nortel does not. The telco needs to offer the service to everyone, otherwise the marketing and operations are a nightmare, and the regulator gets on their case for failing to serve segments of the population. I've been a part of these sorts of negotiations and I can tell you there is no love lost on either side. There is a mutual dependency between them, but it isn't a happy one.
Then there's the matter of pricing. Here we see a common thread with other software products. When Nortel sells a software package to a telco, even though the price is high it is a one-time sale. In contrast, the telco uses that software to generate recurring monthly revenue forever. The manufacturer has had to accept this situation since they in turn have a dependency on the telco - the telco is their sole distribution channel to reaching the consumer market (see the diagram above). When the manufacturers did try to structure software pricing so that they got a piece of the retail action (they called this "risk sharing") it was no surprise that the telcos pushed back, and pushed back hard.
As you can see there is ample reason for friction between these two sets of corporate behemoths. Theirs is not a warm and cozy business relationship. With this knowledge we can now understand why IN matters:
IN was the telco's way of wresting control of service content, cost and delivery from their suppliers, with the ultimate objective of winning a greater share of the profits. To control the phone, your phone, they needed to free themselves from the clutches of their suppliers.
The telcos also had to achieve signalling standards so that equipment from disparate suppliers could talk to each other as their software intelligence level increased in complexity and capability. They and their proxies (primarily Bellcore, since transformed into Telcordia) attempted to use the standards bodies to advance their cause.
All of this took many years. It not only sucked the time and resources of many major corporations, the entire concept suffered from some fundamental problems. Here are what I believe are the worst these problems:
- Expertise: There has always existed some tendency to trivialize the scale and complexity of the technology that comprises the network by those who would wish to exercise control. This has often been true of the telcos. "It's just software," you could almost hear them say. The truth is somewhat harsh to those aspirations. Whatever one may think of the equipment manufacturers, one should acknowledge the depth and breadth of expertise they have developed. For all their own domain of expertise in operating networks and marketing to a vast population, this expertise they did not have. Even when handed the tools they made little progress. A host of proprietary service creation environment (SCE) products came on the market, all of which promised great riches from new services yet failed to deliver more than the basics. Even those were bought at great expense due to technical challenges (see next bullet). Languages like Parlay and JAIN came on the scene and did little better.
Most ended up outsourcing their feature development back to those same companies, but by paying for the show they retained much of the control of those services. - Theoretical challenge: The base code of the switching equipment is, as already mentioned, closed. Unlike Microsoft (as one example), no APIs are provided to enable outside software development. IN was an attempt to define a middle layer, somewhere between raw code, like C, and high-level flow charts, that would define atomic service building blocks and the operations that could be performed on or with them. These would plug into a limited number of "triggers" implemented on the switches.
Anyone with a passing knowledge of formal language theory would likely get intrigued or even uncomfortable on hearing this description. It's a rat's nest of theoretical problems, some approaching intractability. As the difficulty came to be understood, the scope of IN was scaled back. In the North American system, AIN, there were precious few triggers exposed by the switch call model, and even those had to be treated tenderly to avoid catastrophe. Interactions among components was hard enough on its own, and was made immensely more difficult by opening up the software to outside agents with unknown and unpredictable behaviour. The problems remain to this day. - No new features: If you're like me you use the phone to call people and talk to them. You also make some use of additional (vertical) features such as Caller Id, Call Waiting, Call Forwarding (to voice mail), Call Screening, Hold, Conference, and perhaps Calling Name and a few others. What more do you need and are willing to pay for? The answer is, not much. Even when more features are bundled into the service price they are rarely used. Most suffer from being only occasionally used, which means you cannot remember how to use them. Using them is never easy because of the limited signalling capabilities that the network exposes to consumer devices. There are only so many dial codes and combinations of hook-switch flash that a person can deal with. What people want is not more vertical phone services, and they certainly will not pay for them, but rather more communications possibilities, such as those enabled by the internet. There is no market demand for IN-enabled services.
- Consumers are tapped out: We are not looking for ways to spend more on phone service, but rather to do more, or even the same, for less. If a telco does somehow succeed to invent and deploy a new vertical feature that garners some market interest, there is little willingness to pay. Consumers are refusing to spend more on phone calls. They are looking beyond that to smart phones and the internet to deliver something totally new, which they may indeed pay for.
Lastly, VoIP (internet voice) now has enough momentum to finally kill the legacy voice networks. It won't happen quickly but it will happen. VoIP threatens to upset all the telco's plans to maintain control of your phone and protect the revenue of the services they sell. VoIP and mobile will be subjects of future articles in this series.
Posts
No comments:
Post a Comment