content management

The digital conversion is changing every aspect of media production and distribution. And with these changes, the software systems that help to manage and leverage systems and the content they create are becoming increasingly important.

Content management is not new to television or media production. We have always had systems to help us find, store and deliver content. However, because of the physical nature of the media, these systems tended to be physical and manual. Additionally, these tended to be backend systems that were not part of daily production and not depended on for primary success of businesses.

With the advent of digital production and content management systems, we have the opportunity to produce and extend content more aggressively than ever before. CNN has always been successful in moving content quickly from anywhere in the world to air. However, new opportunities such as narrowband to wireless devices and broadband will require dramatically more content. As a result, our systems need to be able to use, modify and deliver this content more interactively. We will need to be able to manage, access and deliver content with greater flexibility. These emerging management tools are the key enabler to our new businesses.

The complication is that media management systems targeted at television are relatively new. Database technology has existed for several years, and workflow and management software has existed in other industries. However, systems that can handle the large rich media files of television and media production are relatively new.

As we look to the future, two factors will help us better align facilities with our systems. First, technology will improve. Solutions already exist and development in this area is in full swing. Software companies, like broadcasters, see the future of rich media and are motivated to be part of the next explosion of content delivery. Secondly, our industry is moving toward a more computing-based environment. Both producers and consumers are adapting to information technology platforms and software-based systems while moving away from dedicated analog hardware systems.

The past few years have been marked by a greater number of initiatives in content management for broadcasters. A growing number of vendors, including Bulldog, Emotion, Informix and IBM, have all listed content management as a key component of future systems. Broadcasters of all sizes, including CNN, have listed content management as a key part of their futures.

As CNN starts its second major project in content management, common questions include: What business logic motivated the projects, how will it change our business, and what does the future look like? While these are simple questions, they lead to complex answers since this is an evolving area. Based on what we've experienced, we will share our views, on how content management will continue to impact production, distribution and the user experience.

The why CNN strongly believes that content is king. To be successful we must produce the best content, manage it efficiently and extend its use more effectively than ever before. We are also striving to be flexible as the Internet changes the rules for our consumers, allowing them to access information via an increasing number of outlets. These new services will put new demands on our production systems and require effective management to deliver more targeted content. It is therefore imperative that content management is a core requirement for our future systems. While we also see some direct return on investment through becoming more efficient with our content and adding ways it can be used, the core return on this investment is our ability to meet these new and changing business opportunities. In short, we think our new production systems must revolve around content management systems to be completely successful.

The how CNN's approach to content management is one that represents both the diversity and centralized nature of our services. While we have always been successful in leveraging a common acquisition for multiple, we do not believe in a one-size-fits-all approach. We fully embrace the idea that CNN's future systems will be made up of many components targeted at particular needs. However, we do plan to extend our centralized approach to our production cycles, from content acquisition throughout production, providing all of our journalists and consumers common access to content. Our goal is to make content easily accessible to all users both inside and outside CNN's firewall.

This approach is represented in two major content management projects at CNN. First, a system we call "MediaSource" handles all incoming feeds and the daily editing processes. MediaSource provides indexing, searching and management of over 1500 hours of low-resolution video, keyframe storyboards and metadata. This system, built with CNN by Informix, SGI, Kasenna and Virage, helps us better visualize video and the production process at over 300 desktops in the CNN, CNN Headline News and CNN International newsrooms. It is scaled and tuned to handle the high volume of daily video production.

Our second initiative involves the CNN Archive and middleware framework for content management. This project targets both the creation of an all-digital system to house the rapidly growing CNN Archives (now at 110,000 hours) and builds a middleware layer for the CNN enterprise to connect content management and digital systems. Sony and IBM are building this system.

In both projects we have consistently maintained our overall content management goals. Together they represent our content management strategy. We will build a system that allows users to get access to all content in a centralized fashion without sacrificing individual needs in production areas or services. Various targeted media management, editing and movement applications connected via middleware will do this. This framework will allow us to meet user requirements, provide a scalable and flexible solution, and remain ready for future changes and growth.

The future The future of digital asset management presents an interesting duality - the highly hyped "convergence" of television and the Internet and the long-term "divergence" of content for delivery on a variety of digital platforms. Convergence has long-term implications for how future "internal" production systems and workflow are organized and operated. Divergence represents the marketplace of tomorrow, or the external applications of content management to satisfy the demands of the next generation of wired or, more likely, wireless consumer.

Internal content management systems leverage the convergence of digital content types to optimized content production systems. Internal systems are valuable for recognizing the value of content as it applies to making decisions on how (and to whom) that content can be packaged and presented. This allows various versions of an asset to be created and associated back to the source data or the parent. In this way versioning is simplified to saving each set of edit decisions or changes, and reflecting those on the source data. This conserves storage media, enables desktop production or publishing, and eases creation of several versions in different formats and data rates to satisfy the demands of an evolving marketplace.

Understanding and capitalizing on the fundamental differences between the context and actual spirit of the content is particularly important for future profitability of media businesses. Obviously, (and particularly true in the news business) the original context of the content is important - it represents and tells the story. But its real value comes when we consider the content in all its various forms. Through the application of databases and content management applications, we can associate content based on its visual and aural content. Various applications are emerging that will allow the search and retrieval of assets based on their visual content, and this will prove very valuable in finding specific faces, locations or shots of specific focal length and composition.

Furthermore, by sorting and cross-referencing the content, we can build collections of data subsets that reveal new value in the video. As the passage of time brings additional value to old items, so the evolution of news stories sheds new value on old assets. For example, the video of President Clinton in a receiving line after his second election would seem to have nominal value. Once it is discovered that the young woman he embraced in that line is Monica Lewinsky, the value of that particular video rises dramatically.

External systems broaden the value of the content to the end user and enable the content to be repurposed in a meaningful way. This allows the essence of the content to be reformatted in both format and context for delivery on a variety of existing and future platforms, such as television, the Internet, WAP cellular phones, PDAs and pagers.

The bottom line is the ability to effectively repurpose the content. By creating and maintaining useful metadata, such as text logs, timecode and standard formats for the exchange of content data between systems, content can be easily and effectively repurposed. It may be repurposed in format, as in transcoding a video presentation from an MPEG-2 file to QuickTime for presentation on the web. It may be repurposing video of a particular news event as a set of stills that can be used in a print publication. Or it may be completely reformatting a news story on a webpage so that it can be received and viewed in a meaningful way on a cellular telephone.

External systems are also necessary to create interactive television systems. ATVEF and other ATV applications will rely heavily on content management systems to understand and react to user profiles and orders, to repurpose and deliver content fitting the receiving device of the user, and to take on an active role in delivering non-real-time content to users as faster-than-real-time files or as stored-and-forwarded streams.

In the next decade, we expect content management systems to evolve into fairly intelligent applications that are not only highly interpretive in managing and providing content but are also conversant with other content management systems. Emerging standards in data exchange formats such as MXF and AAF will continue and eventually include individual user addressing. This is particularly important in creating persistent user profiles, in creating "personal content management" systems and in allowing effective interaction of the user with all the worldwide collections of content.

As wearable computing and other passive, low-cost and small form-factor delivery platforms become available, content provision should become as pervasive as hydrogen. Depending on the available bandwidth for delivery, intelligent systems will anticipate the needs of the user based on where he or she may be (walking, driving, sleeping). Such systems will make decisions as to what content has the highest priority, what content should be delivered right away and in what form, and what content may be of particular use to this consumer based on his or her profile. Personal content management systems will guard the user's privacy as well, refusing content that may be hurled at the user as offers or advertisements from nearby stores, restaurants or media providers.

Repurposing content, analyzing and extracting features important for finding and navigating content, and intelligence to reach out and grab the user are all strategies central to the effective leveraging of content management systems.

In the opening we noted that improvement in technology, coupled with changes in our industry, will better align content management with media production. And while we think it will be a long time before systems required to run our networks and content services are as "off-the-shelf" as word processing, we do believe that systems will evolve to the point that cost and difficulty of implementation will not limit their use. Software that can help manage content will be produced by a large number of companies. These systems should allow for high levels of connectivity, offering a diversified approach in scaling and matching software applications for specific purposes. These systems will also scale from small operations to large, complex enterprises. Most importantly, there will be less concern about success as solutions will be a more standard part of all components.

The key to these systems will be flexibility and scalability. Well-designed systems will allow users to grow and change without major forklift upgrades. They will allow small installations to work effectively with large diverse systems without compromising features or capabilities.

Where to start It is important to have a clear directive or business plan that sells a clear message to all users of targeted implementations. While we don't think we must take a one-size-fits-all approach, we do think it important to develop a consistent strategy that aligns our businesses with the future of content delivery. It is also important to pick a vendor who can either offer a proven system that meets your requirements, or select motivated partners to deliver the solution. Successful partners are those who aim to provide products in this space and are motivated to produce systems that have a life (and viable market) beyond your project. This motivation helps when the project becomes difficult. It also ensures that you can move gracefully from a custom to a more general purpose system that can be more easily developed and more effectively supported.

Content management systems are a vital part of all future systems and we, as content producers and broadcasters, need powerful and effective solutions. The good news is that these systems are available and evolving quickly. The better news is that this is the ideal time to get involved and help shape the future.