Monthly Archives: July 2006
CTAM Summit ’06: Tech Roundup
by Leslie Ellis // July 27 2006
From high definition TV as a broadband video killer, to multiple declarations that user-generated video is the hottest thing going, to Intel’s repeat observation that cable “doesn’t have to die,” the recent CTAM Summit, which gathers cable marketers in one place to talk shop, produced a laundry list of juicy, tech-side tidbits.
Let’s start with this spicy quote from Mark Cuban, CEO of HDnet: “I’ll tell you with complete certainty: No one will be streaming HD video, not this year, not in 5 years, not in 10 years.”
His point: Overall broadband capacity, and especially video-over DSL (digital subscriber line), isn’t roomy enough for high definition video streaming, which requires 6-8 Mbps per stream — even with the most advanced compressors.
And yes, the language of HDTV remains confusing to those consumers wading through the gibberish of aspect ratio and screen type. But from a retailer’s perspective, HDTV is “a steamroller that’s moving very fast & it’s a runaway train,” according to Douglas Moore, EVP/chief marketing officer, Circuit City. As for high return rates on HDTV sets — they aren’t as high as, say, phone answering machines, and “very few people actually bring (an HD set) back to the store.”
Also on the retail front: When it comes to making the decision to buy an HD set, women are as influential, if not more so, than men, said Gary Shapiro, president and CEO of the Consumer Electronics Association. “For some retailers, it’s a shift they haven’t figured out yet.”
Nearly every keynoter at the CTAM Summit mentioned the term “user-generated video” as the new-new thing — including Judy McGrath, CEO of MTV Networks; Shawn Gold, chief marketing officer for MySpace.com; and Intel’s McDonald.
(If you’re older than 30, “user-generated video” is the stuff you capture on your cell phone, and send along to your friends. If you’re under 30, it’s the stuff you capture on a phone or a camera, edit and manipulate to make it better/funnier, and upload for mass-sharing to a web site like Youtube.com or JibJab.com.)
McGrath, for instance, called user-generated content “the biggest trend in the first six months of this year.”
The sweet spot for short-form content appears to be eight minutes, by the way, or at least no longer than 30 minutes. So much for the American attention span.
Self-generated video cropped up regularly in breakout sessions, too. Our favorite one-liner on the subject: “Youtube-styled video is the new black.” (Compliments of Leo LaPorte, on-air personality for TechTV Canada.)
Time-shifted video is now nearly mainstream; location shifting is next. No great surprise, then, that Sling Media’s Slingbox came up in several CTAM Summit sessions.
From the operator side, it’s not as easy as duct-taping a location-shifting module onto a set-top, because “then I’m in a thousand and one conversations about whether or not it’s allowable, in the construct of our commercial relationships” with program networks, noted Mike Lee, chief strategy officer for Rogers.
From the observer side, Sling is cool, useful, and easy to use but not the most beautiful consumer device to grace the living room. “It’s such an ugly little box,” said Molly Wood, executive director of CNET.com.
TechTV’s LaPorte: “It’s the ugliest industrial design I’ve ever seen.”
How to Pick the Winners
Perhaps the most practical bit of marketing tech talk came from Kaan Yigit, president of Solutions Research Group, who offered four attributes to help determine whether a new technology is headed for success, or is just another interesting idea that falls by the wayside.
The four attributes: Things that save consumers time, things that show up in business and life, things with pricing that conforms to the existing value equation, and things that conform to the dominant design.
Example: DVRs save time by scrunching the viewing of a 60 minute show into 40 minutes (sans ads.) Connectivity devices, like Blackberries, cross between personal and professional use. Subscription pricing is familiar; pay-per-song, as one example, is a relatively new pricing model.
As for conforming to dominant design — remember those weird computer keyboards that split into two pieces, supposedly for ergonomic relief? They never really took off, because they didn’t look or feel like “regular” keyboards — they didn’t conform to the dominant design.
Not Dead Yet
Probably the most peculiar outlook at this year’s Summit (besides Freakonomics author Steven Levitt likening cable to crack cocaine) came from tech-side keynoter Donald McDonald, VP/GM of Intel’s Digital Home Group. During his speech, he interspersed two death-related quotes (one by Isaac Asimov, one by Somerset Maugham. Then he closed his remarks with a similarly odd observation: “You don’t need to die.” (Good to know.)
If you’re thinking you’ve seen this Intel scene before, you have. Andy Grove, Intel’s guru, spoke at the same conference, on the same day (July 18), in 1995, and delivered a similar theme: Keep ahead of the broadband competition (which, at the time, was telco ISDN) — or else.
In closing, and as a matter of full disclosure, I assist CTAM in building that necessary but elusive bridge between technology and marketing. I can say with certainty that the amount of tech-side material emerging from this yea’s Summit surprised me — in a good way. Even Intel’s death comment redux was amusing … must’ve been that British accent.
This column originally appeared in the Technology section of Multichannel News.
Why OCAP and the July ’07 Deadline Move as a Pack
by Leslie Ellis // July 17 2006
Maybe this has happened to you. (And if it hasn’t, it will.) You’re talking to someone about what it takes to do an OCAP (OpenCable Applications Platform) launch. You’re pretty sure you were paying attention the whole time. But the next thing you know, the conversation has swerved into the July 1, 2007 deadline that bans operators from deploying digital boxes with embedded security.
Or visa versa: You’re talking to someone about what it takes to meet the July, 2007 separable security deadline. Within three sentences, you’ve careened into OCAP. Here’s a recent example from my notes of a technical session at the recent SCTE Cable-Tec Expo:
“Some initial steps in getting ready to support leased boxes with CableCards involved understanding the headend upgrades needed for OCAP.”
What does a deadline for digital cable boxes equipped with CableCard slots have to do with a promise to the Federal Communications Commission that a middleware layer would start rolling out this year, and continue over the next two or so years?
The range of responses I got to this query, over the past few weeks, perhaps shows how confusing it gets when technology and policy collide.
Here’s a sampling:
“What happened was, as the (July 07) deadline got closer, people got used to saying that OCAP is a requirement. It’s not. It’s a social requirement.”
“We must provide separable security, and everyone I’ve talked to has chosen to use OCAP as the method for doing that.”
“You need an OCAP environment to run DCAS (downloadable conditional access systems).”
Boil it down, and it’s sort of like when you’re planning a major project that will take two years — but part of it has to be ready next summer. The part that unfolds over two years impacts the part that’s due next summer, so you might as well cram both into the short-term to-do list.
Meet the Deadlines
There are two federal deadlines at work here. One is self-imposed: Three months from now, in October, cable’s larger operators will begin deploying network support for a handful of systems to run OCAP, just as they promised the FCC they’d do. They made this pledge to show their unity around one middleware platform for cable-specific hardware and services. (OCAP brings the “and services” part.)
The second deadline is FCC-imposed. It stipulates that by next July, multichannel video providers (that means telcos, too, so watch for exemption requests) can no longer deploy digital boxes with embedded security. Instead, they need to deploy boxes that use separable security — meaning boxes equipped with a CableCard slot, for now, until DCAS is ready.
The two seemingly unrelated projects come together where all major video technology efforts begin: In the headend gear. Headend units need a software upgrade to do OCAP, and they need a software upgrade to enable renewable security on two-way, digital cable ready units (boxes or TVs).
And it turns out that those two software tasks will emerge in the same major software release. So that, in part, is why OCAP and separable security travel as a pack.
Getting Ready for OCAP
There are two camps, when it comes to discussing OCAP field preparations. The first camp holds the “no problem, minor, don’t worry about it” people. The second group, which usually involves those already getting test labs ready, is more serious about the challenge. They describe it as a series of steps.
First, as noted, is a software update to the headend controllers that manage and distribute digital video to receiving devices. In Motorola speak, they’re called “DAC,” for “Digital Addressable Controller.” In Scientific-Altanta’s lingo, they go by “DNCS,” for “Digital Network Control System.” In practice, they work very differently, but as a generalization, they tackle the same tasks.
(Muddying step one for some operators, namely Comcast and Time Warner, is the addition of the forthcoming Adelphia systems into their mix. Comcast, which was 90-ish percent Motorola, becomes 25-ish percent Scientific-Atlanta. Time Warner, which, in its digital video deployments uses mostly S-A or S-A clones, widens its Motorola properties to about a quarter to a third of its total footprint. This means both operators need to bulk up on human resources necessary to understand and ready “the other system.”)
The remaining steps vary fairly significantly from one system (Motorola) to the next (Scientific Atlanta), and as such will be covered in a later column.
All the while, none of the OCAP preparations can disrupt the operations of the legacy devices, already in the field, and incapable, from a resources perspective, of running OCAP.
The good news is, the technologists readying their labs and systems for OCAP tend to use practical language to describe the effort. As in, I haven’t yet heard anyone wearily describe the effort as a disaster in the making, or anything equally draconian.
October is now three months away, and next July is a year away. If nothing else, expect to hear a lot more talk about OCAP, squished together with the separable security, as the year advances.
This column originally appeared in the Technology Section of Multichannel News.
The BackBone People: Goodbye Satellite, Hello Optics
by Leslie Ellis // July 10 2006
Here’s something interesting from an unsung segment of industry thinkers: Satellite-based delivery of digital video material is on the wane. Fiber-based delivery, over owned or leased regional and national networks, is on the rise.
Says who? The Backbone People, that’s who. Who are they? They move amongst us. They own houses and have kids and dogs. When people ask them what they do, they say they work in the cable industry.
Yet they talk differently. They don’t talk, for instance, in terms of how things move, from the headend satellite receiver, over the plant, to homes. They don’t care much about set-tops, or cable modems, or voice adapters.
Instead, they talk about the upper plumbing of the network, north of the headend. They talk in “layers.” They identify themselves by where they reside, within those layers, like this: “I’m more of a layer 2/layer 3 kind of guy, but I came up through layer 1.”
These are The Backbone People. They proudly refer to themselves as “the vertebrae of the industry.” (Nyuck.)
On the operator side, they’re the people who link cities and regions over optical fibers. They’re the people who, when the @Home Network collapsed with such fire and brimstone in 2000, literally airlifted routers and optics stuff to different points in their network’s connective tissue, to make sure cable modem customers stayed up.
On the vendor side, backbone people sell things with oddly nerdy names — like “ROADM,” which stands for “Reconfigurable Optical Add-Drop Multiplexor.” (Backbone people say “ROADM” as a word that rhymes with “modem.”)
This week’s translation aims to uncork some of the curious language of The Backbone People.
Let’s start with those layers. When people start talking layers, they’re referring to a reference model, widely used in data communications, and known as the “OSI stack” — where “OSI” stands for “Open Systems Interconnect.” The OSI stack was determined by the International Standards Organization (the “ISO”), which is a heavyweight in the world of standards.
The OSI stack contains seven layers, each of which defines how some types of data should interact with other types of data. In short, it’s a recipe book for equipment and service interoperability.
Data engineers often remember the layers with a mnemonic. Here’s one I learned (and subsequently cleaned up for print) from a rowdy Irishman at this year’s National Show: “Please Do Not Take Sweet Pleasures Away.” (It was the “s” that I changed.)
In that memory aid, the layers stack up like this:
— “P” is the “physical layer,” also known as layer 1.
— “D” is the “data link layer,” or layer 2.
— “N” is the “network layer,” or layer 3.
— “T” is the “transport layer,” or layer 4.
— “S” is the “session” layer,” or layer 5.
— “P” is the “presentation layer,” or layer 6.
— “A” is the “application layer,” or layer 7.
Generally speaking, those of us who look at things in terms of the headend-to-home routes are “layer 1 & 2” people. People who write PC applications, like firewalls, are “up the stack,” starting at around layer 4. Backbone people are in the middle.
It used to be that “the backbone” of any cable operator grew in one way only: Width. As in, a wide pipe that got wider and wider. Backbones get into big numbers very quickly — an OC-48 link (which equals 2.4 Gigabits per second) grows into an OC-192 link (10 Gbps), and so on. This is already happening.
But what’s also happening is that backbones are growing in strategic depth.
Here’s an example. For five or so years now — ever since the collapse of the CLEC market, which left behind a relative glut of high-speed, metro-to-region-to-metro links, the name of the transport game has been the widespread availability of cheap, Gigabit Ethernet links.
It is those cheap GigE links that drove cable operators to re-plumb their video-on-demand networks, a few years back, to centralize storage servers and stream data, inexpensively, out to hubs and nodes.
It is also those cheap, GigE links that are causing some backbone people to suggest the sooner-than-later end of satellite distribution altogether. Why spend zillions squirting video up into space, and back down again, when it can zip along over glass at considerably less expense?
And this is where the “ROADM” devices come in. Their job is to intelligently deliver bandwidth to a region, or a city, or a town, or a hub, when and as needed — without having to dispatch a person or a truck to that location, to futz with wiring. More on this in a later column.
Boil it all down, and it means this: If you, like me, have spent the last 20 or so years of your life locating cable headends by looking up for the towers or the dishes, you may, soon enough, need a different navigational aid.
Fiber routes are largely underground. Good thing there’s Mapquest and in-vehicle GPS.
This column originally appeared in the Technology section of Multichannel News.