Position paper for the Second W3C Web and TV Workshop Tristan Ferne, Senior Development Producer, BBC Matt Hammond, Research Engineer, BBC Olivier Thereaux, Senior Technologist, BBC George Wright, Head of Prototyping, BBC The BBC is one of the largest providers of TV and radio content online, notably with its iPlayer (http://iplayerhelp.external.bbc.co.uk/help/information/about_bbc_iplayer) offering and involvement in various IPTV and hybrid TV and radio delivery initiatives. These efforts have led us to make a large majority of our broadcast content available on-demand on multiple platforms and we have been a consistent early and strong adopter of Web technologies to provide these services. 1- TV and Radio content on the Web The public mission of the Corporation, to "enrich people's lives with programmes and services that inform, educate and entertain" involves providing quality content to all UK residents, regardless of ability or mode of access. Accessibility, with all its meanings, is crucial to us. We believe in making content accessible on a very large number of platforms and that the best way to achieve this is to make content available through open standards and interoperable technologies. Also important to this question of accessibility is the need to protect rights holders. In the realm of web video and audio there is a common misconception that content protection can be achieved through streaming (because of a misunderstanding about whether streamed content is actually being downloaded to the users' hard drives). Most content providers currently opt to implement content protection through the usage of proprietary DRM systems and closed-source clients (such as Adobe Flash applications). The BBC itself is currently using such methods, a decision we have taken in order to reach as many people as possible in a simple manner while protecting the content we produce and broadcast, but we would see the benefit in having an alternative based on open standards (http://www.bbc.co.uk/blogs/bbcinternet/2010/08/html5_open_standards_and_the _b.html) and our research activities are actively exploring open and free standards for on-demand delivery. The question of how to implement content protection and management enforcement systems in the current Web stack (HTML5 and the video API) is yet to be resolved successfully. The discussion so far in the W3C and other technical fora (see e.g. http://www.w3.org/Bugs/Public/show_bug.cgi?id=10902) recognises the industry's demand for content protection, but with no clear solution in sight. So far there appears to be reasonable agreement on the difficulty of building a content protection client entirely based on an open stack, and that content protection mechanisms may not be added to the specified HTML5 video playback implementation. According to the emerging consensus within the HTML5 community in and around the W3C, the most promising lead may be to provide a reliable mechanism for a client to state whether it supports a specific content protection scheme in a specific video format (see e.g. https://bugs.webkit.org/show_bug.cgi?id=47591), through a standard API method such as CanPlayType() (http://www.w3.org/TR/html5/video.html#dom-navigator-canplaytype). 2- TV and Web (linked) data The BBC has vast experience in using, producing and contributing to the technologies of linked data. As a producer and broadcaster of millions of hours of content it is in our interest to ensure that this content is easy to address, describe and enrich. We have thus had extensive involvement in the realm of linked data and the semantic web. We were heavily involved in the work done by the TV-Anytime forum (http://www.tv-anytime.org/), for which a structured data model to represent programmes and services was built upon XML schema. And more recently the BBC has also been developing RDF ontologies both for music (http://musicontology.com/) and programmes (http://www.bbc.co.uk/ontologies/programmes/2009-09-07.shtml). These ontologies and others are used in various sites built around linked data such as BBC Programmes (http://www.bbc.co.uk/programmes), BBC Wildlife Finder (http://www.bbc.co.uk/wildlifefinder/) and BBC Music (http://www.bbc.co.uk/music). The BBC has also been a contributor to a number of other linked data projects; supporting projects such as URIPlay (http://docs.atlasapi.org/spec/api/), a metadata querying engine for video and radio, contributing code to the Totem media player connecting to URIPlay and generating substantial amounts of original information for Wikipedia and Musicbrainz from our work on BBC Music. Not only a provider of semantic linked data, the BBC is also a heavy consumer of such data, for instance in our BBC News and Sports website (http://www.bbc.co.uk/blogs/bbcinternet/2010/07/bbc_world_cup_2010_dynamic_sem.html). The BBC's interest in semantic web technologies comes from the value we see in both consumption and production of linked data. We look forward to contributing to the future of the semantic web and are eager to further explore needs specific to TV and radio. 3- Beyond "Web on TV" Web-based metadata around TV and radio content is only one of the many ways that web standards can improve the interactivity of and around TV content. We believe there are many ways the Web technology stack could be used or extended to create a better, more interactive and connected television (and radio). HTML-based widgets, for instance, are often mentioned when discussing the adoption of Web technologies by the TV industry. They are, however, only a small fraction of the added value Web standards can bring. While "Web on TV" offers interesting, albeit controversial, prospects, we believe that Web standards, and in particular the family of specifications around HTML5, have a greater role to play in the television of the future. Much of the BBC Research and Development's work in the area stems from the opportunities to use the web and native applications on personal devices, such as laptops and mobile phones, to access and control TVs, set-top boxes (STBs) and other devices. Users will be able to benefit from the many accessible interfaces and systems already available on most personal devices giving those with impairments accessible and familiar UIs that are not limited by the existing UI of the TV. This also opens up an exciting mode of control for all users and the possibilities of second screen scenarios. We are working on web based APIs to enable this, and have developed prototypes for STBs, phones and laptops demonstrating control, synchronised playback of alternative audio, integration between websites and the STB and speech-based interfaces. We are currently working with OEM manufacturers to encourage eventual implementation of these kinds of technologies. We believe a single, standard and interoperable API for these features is desirable, and think that there is room to grow existing W3C work on device APIs to include TV-specific features such as making content portable and ubiquitous and to expand the existing device APIs for audio and video devices to cater for sound and image hardware access, recording and playback control, local storage, synchronisation and data exchange for second screen scenarios, etc. The BBC is happy to see the W3C organise this second workshop on Web and TV and expresses strong interest in the topics and potential work which could result from this workshop. Our context, past and present work lead us to the following three questions as our initial input for the workshop: Q1: How do we reconcile a mission to provide accessible, addressable and interoperable content with the current industry practice of protecting copyright through content protection and obfuscation? The BBC is eager to have other participants share their experience on the issue, and is willing to contribute to a solution which would allow reasonable content protection based on a standard, interoperable web technology. Q2: What are the needs and best practices for standardisation of metadata around TV content? Is it desirable? What work has already been done by other participants in this area? Q3: How can the W3C's existing work on device APIs extend to televisions, radios, set-top-boxes and other audio-video devices?