Cable Takes Action to Clear Last Hurdles to Scaling EBIF

Tony Werner, CTO & EVP, Comcast Cable

Tony Werner, CTO & EVP, Comcast Cable

November 25, 2009 – The cable industry hopes to expedite the push into advanced advertising and interactive TV by streamlining execution in crucial areas related to reliable data transmission and uniform means of matching applications to specific business terms.

While MSOs have made great strides in seeding their headends and digital set-tops with the software components that will maximize the reach of new modes of advertising and interactivity, the pace of implementation will depend on how quickly these remaining issues can be resolved.
“We’ll do many things next year, but it takes awhile to get to larger scale,” says Nomi Bergman, president of Bright House Networks.

Indeed, the scaling issue has become critical now that a growing number of programmers, advertisers and applications developers are buying into the opportunities associated with the industry’s embedding of the EBIF (Enhanced Binary Interchange Format) client software in millions of legacy digital set-tops. “I think we will get scale,” says Comcast CTO Tony Werner, noting his company will have “somewhere north of ten million households with EBIF agents” by year’s end.

But he acknowledges there are “still a lot of pieces and things to work through.” For example, there’s work to be done on integration of specifications and “getting user agents acting in common across multiple MSOs.” As another case in point, Comcast has been rolling out new apps and advertising models on a local basis but now the company must get to where it can implement work flows seamlessly across its national footprint, he says.

Werner and just about everyone else in the industry is confident these hurdles will be cleared. “We’ll get there, and I think it’s going to be huge,” he says.

That confidence is buttressed by initiatives MSOs are taking individually and collectively through the auspices of CableLabs. “The technology has been specified; the set-tops are out there,” says Paul Liao, president and CEO of CableLabs. “But now the hard work of getting the networks ready is about to be underway.”

Liao makes clear he wants CableLabs to be “much more involved in some of the practicalities of these deployments.” That entails reaching agreement on best practices and “really understanding how to get the network up and going so we truly have a national platform,” he says. “Basically, this is a massive data system. Every router has to have all the back addresses, switches set correctly, and all that has to be done. And once that’s done this thing will go like greased lightning.”

From a pure engineering standpoint the “pipe cleaning” now underway is the last remaining hurdle to scaling EBIF. “EBIF applications are bound into the programming streams from the programmers to the cable headends,” notes Don Dulchinos, senior vice president for advanced platforms and services at CableLabs. “Those signals have to traverse groomers, transcoders and other processors, and you need to be able to verify they went through. You need a clean path.”

On the business side another big piece of unfinished business is the establishment of a uniform process for making sure a particular ad or application is tied to a specific business deal. “This has to be automated,” Dulchinos says. “We have to set rules and make sure they are caught and read and that execution is verified back to the programmers.”

CableLabs has a working group focused on this piece, he says. “We’re writing it up and documenting performance to ensure that applications match business arrangements with MSOs wherever those apps are running,” he adds.

In mid November CableLabs hosted the industry’s largest and most comprehensive Advanced Advertising Interop to date involving 24 suppliers along with the Society of Cable Telecommunications Engineers, Canoe Ventures LLC and observers from cable, programming and other entities. The week-long event served to underscore that the industry’s efforts to standardize formats and interfaces under the Advanced Advertising 1.0 Specification jointly developed by CableLabs and Canoe Ventures will provide advertisers the scale they’re looking for.
“The great turnout shows the increasing momentum behind the MSO and Canoe efforts to adopt standards that allow for the delivery of new ad forms to a national footprint,” Dulchinos says. End-to-end interoperability was demonstrated for a variety of ad formats including interactivity (such as voting and polling and request for information applications), ad insertion in VOD content, ad insertion in linear content, telescoping and graphic overlays on VOD content. Some of the formats were enhanced with addressable advertising, which taps into the platform’s ability to aggregate customer response information through a common measurement format without exposing personally identifiable information.
While Liao says it appears CableLabs will not have to develop a new version of EBIF thanks to the ability of technicians to work out inconsistencies in user agents that have cropped up in interoperability tests, there likely will be a need for new specifications that complement the existing platform. These won’t be essential for getting the current wave of EBIF-based ads and apps off the ground, but they’ll be needed “if we really want to make use of the fact we have addressable advertising and things of that nature so that we can extend SaFI (Stewardship and Fulfillment Interfaces) applications [to] take maximum advantage of this platform,” Liao says.

Doubts as to the ultimate value of cable’s advanced advertising platform to programmers, advertisers and MSOs from a pure monetization and consumer value perspective are rapidly fading as ever more strategies are tested in the field. “We’re moving through a phase of bold programmer experiments with interactivity,” says Peter Low, president and CEO of ITV software supplier Ensequence. “The results are better than anticipated.”

For example, interactive ads are engaging three to eight percent of viewers who see such adds for anywhere from three to six minutes, Low says. “On the content side,” he adds, “when there’s an interactive enhancement 15 to 20 percent of people will use it, and we’re seeing numbers soar in some instances to 50 percent.” Ratings for a show with interactivity versus the same show without such apps are jumping by five to 20 percent, he adds.

Low stresses these are early days when results, as good as they have been, are likely impeded by the fact that interactivity has yet to become a routine part of the viewing experience. “When interactivity becomes a core piece of the way networks make TV and creative departments are thinking about it all the time, the numbers will go up dramatically, because these people know the interests of viewers,” he says.

A%26E Television Networks has been working with Ensequence in some of these trials. Mark Garner, senior vice president for distribution, marketing and business development at A%26E, confirms Low’s assessment of the appeal of interactivity, adding that the network is finding that when interactive apps are involved people are staying “tuned in for longer periods of time.” As for interactive advertising, “We’re seeing a lot of engagement – ten times greater engagement than you see on our broadband outlets with contextual click-throughs,” he says.

Werner, too, reports that early results with the apps Comcast has introduced on the EBIF platform are very encouraging. These generally fall into three categories – advertising, extensions to electronic programming guides and unbound apps, where the apps, such as local weather or traffic reports, run as text overlays on request from viewers, irrespective of what programming they’re watching.

In the advertising category, Werner cites one referred to as “Ready, Remind, Record” as having strong appeal for both advertisers, in this case programmers, and viewers. The programmer promotes an upcoming show with a message to the viewer telling when the show will appear and asking whether the viewer would like a reminder to appear before it airs and whether they would like to have it recorded on their DVR. “It’s clever,” Werner says. “It’s one that consumers will like and one that the people with the programmers will very much like.”

“We’ve also got an RFI (request for information) app set to run with advertisers,” he adds. “[Comcast ad sales unit] Spotlight is selling that, and it seems to be working pretty well.”

An example of a guide extension is a “More Like This” app, where a banner appears when a viewer clicks on a particular channel asking whether the viewer would like to know about other programs that are like the one just chosen. “This app has been developed but it’s not out yet,” he notes.

One that has rolled out with what Werner describes as “remarkable” results is an on-screen click to buy app running with the Home Shopping Network. He also notes Comcast is now offering caller ID on the TV screen in systems running on Motorola set-tops.

But for all the effort behind EBIF, Werner makes clear the long-range strategy remains tied to the more advanced tru2way platform, a Java-based middleware that runs on set-tops that have the processing power to support much more graphically and functionally rich applications than can be done with EBIF. “We want to put all our [guide development] horse power behind tru2way, which we’re doing,” Werner says. “Next year most of our new boxes will be tru2way, and they will get our ultimate guide with a much more modern look and feel.”