Whatever Happened to Sprint’s ION?

More than three expensive years after Sprint Corp. (NYSE: FON) began to sing about Integrated Online Network (ION), its answer to converged broadband, the operator has changed its tune from one of loud hype to the quieter sound of euthanasia.

Under financial duress, the Kansas City, Missouri-based operator decided in October that $3 billion was quite enough to spend on the failed project (see Sprint Announces Results, Cuts). Its decision to scrap the service will cost the company a further $2 billion in cash at least, several thousand employees, and who knows how much in terms of damaged customer relations. It is currently hoping to disconnect the less than 4,000-strong customer base it managed to attract to ION by December 31.

So how did Sprint get involved in such an embarrassing fiasco? At best you could say that it raced ahead of its peers with a bleeding-edge story. However, the operator was so keen to get the first-mover advantage that it didn’t consider the pitfalls of pushing a service to market ahead of the technology needed to support it.

On its simplest level, back in 1998 when convergence and VOIP (voice over IP) were still hot buzzwords, Sprint forged its ION plan in the hope of achieving two things. The first, and apparently the most expensive aim, according to Sprint spokesperson Robin Carlson, was to converge all of Sprint’s services – including Asynchronous Transfer Mode (ATM), Frame Relay, and voice – onto a single Internet Protocol (IP) backbone. This is the kind of project that operators like Cable & Wireless PLC (NYSE: CWP) are now hoping to achieve using Multiprotocol Label Switching (MPLS) (see C&W Announces Its MPLS-Enabled OC-192 Trans-Atlantic Link). Sprint did not use MPLS.

The operator also wanted to offer business and residential customers high-speed Internet access along with up to four phone lines and voice mails and a fax line over a single local access line and under a single price plan. The service supported simultaneous telephone calls and Internet access.

In 1999 Sprint even bought fixed wireless MMDS (Multichannel Multipoint Distribution Service) licenses with a view to offering a wireless version of ION. The brave idea was to bypass the troublesome incumbent local operators in all of the markets where Sprint held spectrum licenses. Of course, fixed wireless never became a platform for ION, not least because to this day there are no voice-enabled products available for the MMDS spectrum bands. Sprint says it’s still holding out for the next generation of voice-friendly fixed wireless. It could decide instead to use the airspace for mobile services, but that’s another story.

ION for business customers was launched in a limited number of markets in the middle of 1999, about a year after the plans were first unveiled. The consumer service followed in November 1999.

What’s the problem?

The ION service that actually hit the streets included high-speed Internet access based on either DSL or T1 lines. The majority of ION customers were consumers or home workers using the DSL version. The operator used ATM as the underlying technology for its voice and IP access options for businesses – and the consumer voice offering had reliability problems.

ION was powered by off-the-shelf Cisco Systems Inc. (Nasdaq: CSCO) and Nortel Networks Corp. (NYSE/Toronto: NT) transmission and access equipment. An integrated access device (IAD) dealt with traffic management.

According to Giga Information Group Inc. research fellow Lisa Pierce, this is where the problem lay: “Sprint tried to use off-the-shelf Cisco and Nortel equipment to offer a revolutionary managed service. If you want to offer a service of that sort it requires a great deal of customization. Off-the-shelf equipment won’t work. I’m not surprised customers said they had problems with voice.”

This lack of customization meant the service was not compatible with a range of incumbent carrier services including ISDN, inbound and outbound circuit switched telephony (PSTN), and inbound 1-800 services, Pierce added. For example, the business version of ION didn’t support local calls. The fact that ION was neither backward compatible nor equipped to support a guaranteed level of service hiked the prices up for customers, as it forced them to buy additional services and employ their own staff to manage them.

None of these inadequacies sound like very good reasons to pay the premium that Sprint charged for ION. The price ranged from between $150 and $800 a month for consumers, including installation costs. The operator did not list its T1 business prices, presumably because it had to negotiate each deal.

From Sprint’s point of view, the four truck rolls necessary to install the consumer DSL service made it a “ financial white elephant,” according to Pierce.

After struggling along since 1999, Sprint finally admitted in June this year that it was suffering some “voice stability” problems. In light of the poor takeup of the service and the technical pain, the company finally began to question the validity of the whole ION project. In a last-ditch attempt to save the service, the company said that it was testing the possibility of bolstering consumer VOIP with ATM. The operator expected to have a clear result from this trial by September but ultimately balked at the cost.

“We wanted to get as close as possible to five nines reliability, but ATM for consumers would have been costly to scale up,” explained Sprint spokesman Mark Bonavia.

By the time Sprint announced its decision to scrap ION in October, cutbacks had become more important than technology or services. After the events of September 11 the already accelerating decline of the telecom market had become painfully obvious, and the operator scrapped the service – along with 6,000 jobs – in order to show shareholders that it knows how to save money. The company says it has no plans to develop an equivalent to ION.

What if Anything Will Replace ION?

In mid December, Sprint sent ION customers a final notice. Before mentioning that it’s still hoping to sell nationwide long-distance voice services and DSL in selected local markets, Sprint politely informed customers that they could find contact details for alternative voice and data providers in their local telephone directory. Although the operator is offering credits of up to $600 for customers, it doesn’t seem to be rolling its sleeves up and offering hands-on help with migration.

Since the FCC and state regulators need to approve Sprint’s termination plan, the operator anticipates that it could take up to a year to disconnect the last ION customer. Presumably customers are already looking for alternatives rather than sticking around to watch the lights go out.

However, there is no exact equivalent to ION in the U.S. and there are few high-level options for businesses and consumers that still believe in the dream of VOIP and convergence.

AT&T Corp. (NYSE: T)and WorldCom Inc. (Nasdaq: WCOM) are offering managed VOIP services. However, although she rates AT&T’s INCS service ahead of ION, Pierce warned that it also lacks service-level agreements. On the other side, WorldCom Inc has recently been very hesitant to talk in detail about its session initiation protocol (SIP)-based VOIP service, even though it was due to kick off in the middle of this year.

Unless they go for a cable telephony service, which is not available everywhere, the only option for consumers is to buy a DSL subscription from their incumbent telephony operator or one of the few data competitive local exchange carriers (CLECs) that are still in business. However, although these services allow simultaneous voice and data access, they don’t include the extra phone lines that ION delivered.

For converged data and multiple phone line packages consumers and small business will have to wait around for voice over DSL (VODSL). This long-emerging technology can support up to 16 voice lines alongside one DSL data connection.

Although several equipment vendors – including Alcatel SA (NYSE: ALA; Paris: CGEP:PA), CopperCom Inc., and Jetstream Communications –are trying their luck with VODSL products, it is not clear who will offer the services. The incumbents would probably sooner gouge their eyes out than offer VODSL, since they already run voice and DSL on the same lines. CLECs – including Covad Communications Inc. (OTC: COVD)and Rhythms Netconnections Inc. (Nasdaq: RTHM)– showed an interest in DSL last year. Rhythms died earlier this year and Covad, which hoped to offer the service by the middle of this year, is too cash starved and distracted by the need to survive. So far, only very small operators have announced plans to launch VODSL.

ION represents the end of an era. Until and unless WorldCom and AT&T and others begin to use SIP to offer interesting and reliable corporate VOIP services, it is unlikely that convergence will enjoy the glamour it promised towards the end of the last decade.

— Sinead Carew, special to Light Reading
http://www.lightreading.com