Controlling The Phone - Part 10: Standards

In the final installment of this series I need to talk about standards. At first glance this may seem an odd topic for this series since standards, one might think, are a consumer benefit, whereby you can go into a store and choose from among many competitive products that support some prevailing standard. Examples include telephones, modems, routers, Wi-Fi and Bluetooth devices, televisions, DVD drives, and so on. These are indeed good things for the consumer, and all are enabled by standards which enable competition in the product marketplace. However, standards do have a dark side.

In most cases, standards are followers, not leaders. By this I mean that, first, a new technology rolls out, then there is increasing chaos in the market as competing manufacturers produce variants with differentiating features. Finally sanity is restored by narrowing the field to one or at most a few acceptable variants. These final versions are the result of sometimes ugly negotiations that necessarily disadvantage most of the players in the industry. These negotiations can only succeed when the plethora of non-interoperable products causes consumers to stop spending. In other words, when there is money to be made, the industry will normalize.

There are endless examples of the above, from video tape to DVD, from low-speed data over phone lines to DSL and optical fibre, CMDA to GSM, and even cars and gasoline grades. The common thread is that when the initial market acceptance of a new technology is choked by incompatibility, and therefore stalled sales growth, the stakeholders will negotiate standards.

Sometimes standards lead market success. Examples include TCP/IP, email, and the web. More often, it is a sign of doom when this occurs. Consider ISDN, IN, and the Clipper chip. The reason is that when the industry is playing at the standards game at the start of a technological initiative it indicates that they are not innovating and rolling out products to market. Money must come first. When there is no market, you will find that the major manufacturers only participate in order to placate their major customers, who are typically the drivers of the standardization drive. I personally spent some time doing this with ISDN and IN, and let me tell you it was a hopeless affair. The amount of time and energy wasted on non-product standards was considerable and ultimately wasted.

In contrast consider V.90, the 56,000 bps dial-up data standard that got done in months when the market was crying for the increase in speed. Customers were waving money in the air, ready to spend at the first sign of a standards-based product that would work for all ISPs. Manufacturers were even selling pre-standard products with the promise that the modems would be upgraded for no charge once the standard was settled, all to lock in customers. That is how standards work when they work well.

Let's now take a walk on the dark side. In telecom, both service providers and manufacturers see distinct advantages to proprietary products and services in a competitive environment. It is a way to lock in customers. For example, if you have a feature-phone that works on a Mitel PBX system you can bet that it will not work on any Nortel or Lucent business system. To switch to a competitor's PBX requires replacing all of the phones.

This cost of switching is so high that there is a lock-in effect. Cellular service providers try to do the same with the phones they sell, by incorporating components that ensure the phone will only work with their network. The pricier the phone, the greater the lock in. The same has held true in the telecommunications infrastructure market where manufacturers like Siemens and Nortel would aggressively market switching and transport equipment that is non-standard, with the aim of locking in the carriers once they have investment millions or billions of dollars in the initial roll out.

Carriers, for their part, would then attempt to intimidate their suppliers to comply with new standards for future sales. The losing suppliers would, naturally, enthusiastically embrace the standards in the hope of displacing their competitors. Siemens played this role with ISDN back in the early 1990s, as Lucent did with IN throughout the 1990s. SONET, ATM, DSL among others had their own battles. The competitive and dirty tactics that can be witnessed by sitting in on these standards meetings are quite eye-opening. It is a nasty and dirty business, and not for the faint-hearted.

Similar tactics are pursued by governments and their agencies, the telecommunications regulators. The countries that were home to the major multi-national equipment manufacturers regularly enshrined standards in the guise of regulations, and which unsurprisingly reflected the specifications of those same manufacturers' equipment. Thus we saw things like 4B3T line coding in Germany (Siemens) rather than 2B1Q for ISDN, T1 and E1 for digital carrier systems, and BTUP in the UK rather than ISUP for telephony signalling. There was the old joke within Nortel about the international telephone gateway product, DMS-300, being little more than a giant protocol converter for all the incompatible national trunk systems.

The ability to lock down protocols to continue this practice today is much lessened because of the obsolescence of the older generations of technology, the growing need for reciprocate to protect export markets, and the rise of TCP/IP and the internet. There are today fewer means and reasons for regulators to step into the fray. Carriers do still play their games but to lesser effect.

For example, anyone who has attempted to sell VoIP equipment to a telephone company knows that they have abdicated any role in setting standards. Instead they demand that your product demonstrate compatibility with other equipments they have already selected and installed. They do this knowing that the VoIP market is fast-moving and innovative, and there is no benefit for them to dictate terms - it would only impair their ability to change in the future. This has led to ecosystems of suppliers, usually led by a major player like Cisco, with a certification process for interoperability testing. There are also loosely-coupled industry associations that do the same, and in one case a centrally-controlled and operator funded institution for one industry sector: CableLabs for the cable industry (PacketCable).

What standards really come down to in the end is simply a matter of compatibility and predictability. These days there is far less slack in the telecom product pipeline to allow the gamesmanship that typified the standards process in the past. Delay means lost opportunity, which can make or break many companies. This is a healthy development. Of course it is possible that at some time in the future the industry will again reach a plateau with operator-supplier market stasis, but for now the need to make money dominates. Today, standards promote rather than impede business and delivering innovative telecommunications services to consumers. This is as it should be.