Assessing the 'Year of Open Cable’


As it’s been north of two years since this column last delved into the Open Cable Applications Platform (OCAP), and since the cable majors intend to launch OCAP over the next two years, starting this October, it seems timely to dive back in.

A new rack of details emerged last week in a webcast hosted by the Cable & Telecommunications Association for Marketing. Disclosure item: I delivered a brief on what’s what on the subject, but the real goods came from the other panelists: Joan Gillman, of Time Warner Cable, brought the cable view. Jonathan Bokor, of The Walt Disney Internet Group/ABC, delivered a “programmer’s wish list” for OCAP. Don Dulchinos of CableLabs moved the topic forward by describing some of the extensions scheduled to latch onto the specification — some of them by this fall.

In case you missed it, this week’s translation covers the highlights. For more, prowl CTAM’s Web site ( for the archive.


To “level set,” as tech people say, OCAP is a category of software known as “middleware.” It wants to sit above the little embedded operating system, and below the applications, in digital set-top boxes or cable-ready consumer-electronics devices (high-definition TVs, to start.) It’s nine years in development, but all signs point to sizable rollouts in ’07 and ’08.

OCAP matters for program networks because it offers a way to cut multi-operator deals for interactive stuff that lets viewers participate in shows, often in a way that affects the results of the show — like voting along with Dancing With the Stars. It matters to cable operators because it lets them normalize the dozens of makes and models of digital boxes in the field, each with different innards. By abstracting the many differences between boxes, new applications can go up more quickly, and with way less futzing.


The cable view on OCAP is this: It’ll go into newer, higher-end boxes, like those that do digital video recording and HDTV. For the existing base, there’s the OCAP subset known as ETV, for Enhanced TV (see the Sept. 5, 2005, translation).

Whenever someone starts talking about ETV, the term “EBIF” (pronounced “ee-biff”) isn’t far away. It’s another doozy of a tech term that stands for “Enhanced Binary Interchange Format.” That’s code for the code that makes interactive stuff work on legacy, resource-skinny boxes.

But what are the apps? Last week, Time Warner Cable launched a voting application with NBC’s Last Comic Standing, with plans to extend the service into 12 other markets. It already offers a voting/polling application on New York City news channel New York 1. So far, both applications are proprietary, but both are slated to be ETV- and OCAP-compliant.

Most of all, operators want more applications — especially two-way applications, because the dish guys can’t do two-way very easily.


Programmers wish differently. First, they want a heavier foot on the gas pedal. Faster OCAP launches, in more headends, to more set-tops and consumer electronics devices.

Second, they want ways to be proactive about ad skipping, which is happening everywhere there’s a DVR. Specifically, they want to try out things like “telescoping.” That’s where you’re watching the Mini Cooper ad, and it’s of interest to you. You click (that’s the telescope) to watch a longer video about the vehicle. The clip is stored on a VOD server.

On the authoring side, program networks seek an “open source applications template,” presumably from the vendor community and blessed by CableLabs and its member cable companies. The template is precertified and pretested to be good to go.

Over time, of course, programmers want to apply creative strokes that make their interactive efforts as irresistible as possible — but in the beginning, they want a way to write things quickly, with an assurance that they’ll work across cable geographies and set-top landscapes.

Lastly, and perhaps most illuminating, is the desire by program networks for a central testing facility, to make sure their interactive efforts work.

What’s illuminating about that desire is the void it uncovers. The technology is essentially ready. Headends are gearing up. New applications (not re-coded versions of the guide, etc.) are emerging, if slowly.

But nothing yet exists to assure operators that an OCAP app won’t bring anything to its knees, or to assure programmers that their application really will work on all available OCAP devices.


In interactive TV circles, the “chicken and egg” expression often pops out. Programmers don’t want to throw cash at anything interactive until they’re sure there’s a big enough box base to run it. And operators with a big enough box base want applications, before they get in too deep.

There’s a difference this time around. Practically speaking, it’s what’s made OCAP such a dull topic until now: All the operator work, so far, is in re-coding existing applications, like the guide and the on-demand ordering system.

But the very fact that applications are being recoded to run in an OCAP format shows some skin in the game.

So does the work to get headends ready by this October, and on through 2008.

That’s why people are starting to say things like “this is the year of OCAP.” Practically speaking, this is probably the year of the start of OCAP. OCAP will be 10 when it gets its “year.”

Stumped by gibberish? Visit Leslie Ellis at