Even though HTML5 is still evolving, creating user interfaces (UIs) using the Web-based markup language are perfectly feasible today. That said, cable companies engaged in developing HTML5-based UIs are certain to be unnecessarily duplicating efforts.
Designing a UI was simple when the only display a cable operator had to worry about was the TV (and even then, the best UIs were … adequate). Now, service providers have to be concerned with UIs that not only have a similar look and feel across multiple platforms, but are also able to support apps with functionality that cuts across multiple screens. Comcast’s X1 (formerly Xcalibur) is a case in point.
At The Cable Show session “Your wish is its command: New possibilities for technology interfaces,” Comcast’s chief scientist of metadata platform and search services Amit Bagga demonstrated an app for companion devices that TV viewers can use to enter – by keyboard or by voice – searches for content.
The way the app operates would be familiar to those who’ve used Google’s voice input or Siri on Apple’s iPhone. Bagga showed the app responding to voice commands such as “What baseball game is on tomorrow at 7 o’clock?” – with the app responding by delivering the navigation guide listings, in this case for baseball games being broadcast on that day and time.
Navneeth Kannan, senior director of systems engineering for Motorola Mobility, said the industry will inevitably have to choose a browser to support on the set-top. The case for an HTML5 browser is that it is already familiar to developers, it can be made to work with OCAP and it will enable easy UI revision. The challenge is that HTML5 is still in development, and it doesn’t have set-top-specific support features yet.
“Even how to turn on the front-panel LED, there are no standards for that yet,” Kannan said. “We need changes to support platform-specific features.”
There are nitty-gritty details to consider, Kannan said. Graphics resources need to be available, not only to the browser, but also to the platform – for closed captioning, for example. There are considerations for memory allocation and management, as well as for widget authentication and storage. The industry must still develop stronger support for EBIF.
Despite the work that has to be done, though, Kannan said: “This is real. We’ve ported three different browsers. We can do it now. HTML5 apps on a browser are feasible now.”
(In a separate session, Kurt Hoppe, director of smart TV innovation and alliances at LG Electronics, said LG and Cablevision have tested an HTML5-based app).
Michael McMahon, Charter Communications’ vice president of Web experience and application strategy, said that part of the excitement about HTML5 is that HTML makes it possible to write applications once and then do only minor tweaks for other platforms, so that you end up with a minimal code base. The problem is that there’s no such thing as a TV-specific, reusable framework.
“Everyone has their own code supporting every different tablet and smartphone and tweaking those for each iteration of the device. We don’t need to reinvent everything every time,” McMahon said. “It is possible to build a core framework.
There are other areas of commonality, but they have to be coordinated, McMahon said. “CableLabs to me is the obvious answer to this.”
Simon Parnall, vice president of technology at NDS, talked less about user interfaces than about how display technology is likely to change. In that context, UIs will have to evolve with them.
For nearly a year, NDS has been talking about how within five years, it will be likely that displays will break out of the 16:9 rectangle. Light, flexible, wall-mount displays might be small, or huge, and could be tiled in all sorts of geometric shapes, accommodating multiple windows or views that would be seamlessly integrated.
“Science fiction has been talking about this for years,” Parnall said, and he noted that the display in the lobby of Comcast’s headquarters in Philadelphia is an early example.
The interface will have to change. Viewers will need to control the focus on the main item, and to do that, he suggested an immersion dial that makes the focus more or less prominent.
Ben Weinberger, CEO and co-founder of DigitalSmiths, wrapped with discussion of what his company calls time-based data. The concept is to go beyond the current concept of metadata, so that even individual frames of video can be tagged, by anybody, with relevant data.
The idea is to go beyond being able to command your TV to “find ‘Animal House’” to saying things like, “Find me a picture of John Belushi in a toga.”