DOCSIS 3.0: There's a Reason for the Bronze Age

It turns out there’s a really good reason why MSOs are pushing CableLabs to move DOCSIS 3.0 to the starting gate without the normal warm-up period. Surprisingly, if you’re to believe the industry’s top tech minds, that very good reason has nothing – OK, maybe a little – to do with those competitive offers the telcos are hawking like street peddlers at a summer carnival.

"DOCSIS 3.0 is a logical evolution of where we’re going," said Steve Craddock, senior video president of new media strategy at Comcast and an early morning panelist during a breakfast sponsored by Communications Technology and Texas Instruments at The Cable Show in Las Vegas.

By the way, what’s with "The Cable Show?" Didn’t that thing used to be "The National Show?" That’s like the telcos calling their annual gathering next month the "Twisted Pair Show."

Anyway, the speed with which cable is pushing the 3.0 specification is revolutionary. Compatible products should be in the field by the middle of next year, and by "2009 it’s all DOCSIS 3.0 going forward," Craddock said. The goal, he emphasized, is evolutionary even if the movement is revolutionary. The final gold standard of DOCSIS 3.0 – which will follow a bronze standard that emphasizes downstream channel bonding and a silver standard that adds channel bonding to the upstream – should make it possible for cable to hang in there with existing HFC infrastructure and reasonably migrate to IP for the cost of "a couple billion dollars" to blanket 99 percent of the country compared to the multiple billions Verizon is spreading to cover far less territory with FiOS. Who needs it? It’s not like cable needs 100 Mbps downloads or, in the second phase, equivalent symmetrical speeds. It doesn’t need to add boodles – that’s a technical term – of more IP addresses via IPv6. And it doesn’t need to amp up security and reliability to new levels of encryption favored by an increasingly paranoid government.

True enough, Comcast is running out of IP addresses and needs more, if not a boodle or googles of googles more. And if operators want to sell communications services to the government (or government contractors or security-minded businesses) they’d best adopt the latest cryptography. The point is that these DOCSIS 3.0 features, including the very high throughputs, are highly desirable, if not urgently necessary.

"We’re not dying for this coming out tomorrow," said Jay Rolls, vice president of technology for Cox Communications.

This makes the luxury of having it come out faster than the original 2009-2010 time frame all the more delicious. When it arrives, there will be plenty of uses on both sides of the ocean.

"If you can do 100 Mbps symmetrically, you can do some things that are really cool," said Doug Semon, vice president of technologies and standards for Time Warner Cable.

Along with the other DOCSIS 3.0 features, you can, for instance, tighten security and enhance reliability for business customers and attack a monster audience of users that falls between the cracks of big enterprises and SOHOs. The DOCSIS 3.0 MAC protocol gives MSOs serving business customers so much flexibility that "it’s almost like dumb pipe (telco) vs. smart pipe; it could get that dramatic," said Rolls. More IP addresses Comcast needs to move to IPv6 because "we do consume an awful lot of IP addresses," said Craddock.

Of course, there’s also the migration to IP and the opportunity for European operators like Telenet and UPC to fend off increasing competition from telcos delivering DSL.

"I love fiber, but I hate digging," said Jan Vorstermans, executive vice president of technology and architecture for Telenet. DOCSIS 3.0, he said, will let his company squeeze more out of its existing bandwidth so he might even be able to deliver IPTV in competition with a growing number of telcos offering video services in the Old World.

Those are the immediate goals and the reasons Comcast pushed CableLabs into using "temporary qualification waves." The bigger picture is yet to be painted. Services such as home security might climb out of the hole in which they’ve been tucked for ages and become viable offerings.

"Home security … has been called one of the real killer aps because it’s killed every company that’s tried it," Craddock said. With DOCSIS 3.0’s bonded upstream channels, "it becomes a very high margin business," and "you can really change the economics of that business."

The gold standard DOCSIS 3.0 – a goal that every vendor must reach after 2009 – is a specification so complex that it’s more like moving from DOCSIS 2.0 to DOCSIS 5.0, said Semon, who noted that it will enable "fascinating niche areas" and deliver a "high rate of return for niche ancillary services."

It will also, of course, help cable compete with Verizon and AT&T and others that are noisily entering the triple-play space. But that’s not enough reason to push the certification process. A long time coming Verizon’s FiOS, Craddock said, "will happen, but it will take a long, long time. FiOS is something my grandson has to worry about, not me. The AT&T architecture, I’m not even going to touch."

In the meantime, cable "can use DOCSIS 3.0 offensively … grabbing the high ground early. That is something the LECs cannot threaten … for many, many years," Craddock continued. "We’ll have 3.0 deployed significantly ahead of Verizon FiOS." – Jim Barthold