Content management means different things to different people. But while many would say that it is largely about locating content — “it is all about the metadata” is a commonly heard statement — I would argue that it is much broader than that. If a system is going to be truly useful, it has to manage not just the content but also the processes that use that content.
Today, broadcasters and content owners should be considered business entities, seeking ways to get the maximum value out of their assets by monetizing content over a range of platforms. On that basis, there are three central elements in what they do:
- Media (which is not just audio and video, and not necessarily in digital form);
- Workflow processes;
- Resources, technical and human.
Precisely which workflow processes will depend upon the nature of the individual business. A broadcaster or playout center, for example, might have as its fundamental structure:
- Accept a tape;
- Understand its content;
- Digitize it;
- Prepare and process it for delivery;
- Enrich it, for instance with subtitles or other access services;
- Ensure it meets the right quality standards;
- Ensure it is ready for distribution across one or a number of platforms, at the right time with the right associated information.
As you will appreciate, a high-level view of a workflow such as this can be broken down into individual operational steps, each of which requires access to content, and technical and human resources to make them happen.
In the traditional broadcast environment, this has been handled by a silo model. An incoming tape was entered into a database by a clerk and then put on a cart with other tapes to be delivered to the QC suite. There, one of a team of expert watchers checked it for technical standards. Then it was put onto another cart for delivery to the library, which sorted the tapes into transmission order for the playout suite.
This model is operationally inefficient: The QC team could be sitting idle one day and then be swamped with new accessions the next. The late delivery of a piece of content means that the standard process has to be overruled by an exception that involves carrying a tape around, asking individuals to push it to the top of their priority list, and so on.
Like all linear processes, it is also inherently inflexible. The process happens in a straight line, and adding any further requirement, such as producing additional versions for different delivery platforms, has to be added as a series of branches, each with its own linear processes.
The file-based environment
These arguments are familiar and are at the heart of why broadcasters are migrating from a linear system based on audio and video tapes to one in which all content is ingested into a file format at the start of the process. Computer files inherently need a means of search and location; you cannot look for a label on a tape box. There are those who would say that this is “asset management.” I think that is just a database and is not adding any value to the business.
An asset management system has to know about the content, the metadata, the process and the resources available to undertake the process.
With a simple database, you probably know what content is available. But if you do not know if you have the rights to use that content and on what platforms, or what form the content is currently in and where it is, then the content is not an asset; it is a liability, simply taking up storage space and not contributing. And if you do not know how much it costs to process the content from where it is to where you want it, you cannot make business decisions on whether to reuse the content.
Note that in the previous paragraph, I made the point that you need to know the current form of the content and its location. It is an easy mistake to think that in a new file-based environment all the content will be available online more or less instantly. This is not true, and it has to be part of the information you draw from the content management system because you have to make business and operational decisions on that information.
Most obviously, it may not be online but in an offline data tape archive. Transferring it to the online disk array will take some time, even if there is no queue for the data tape robot. The basic laws of physics apply here: An HD clip will take longer to transfer than an SD clip, or absorb more of the available bandwidth (costing more in chargeable system overheads), and that may affect your decision when choosing content.
More significantly, unless it is a brand-new greenfield business, much of the content, and probably the vast majority, will not be within the file-based environment at all but on shelves of tapes, film, discs and other media. When the idea of asset management for A/V content was first conceived, many broadcasters rejected it as impractical because they saw that digitizing all the archive would take far too long and cost far too much.
You have to turn that idea on its head and acknowledge that sometimes the clip you want will be on a shelf. The good asset management system will tell you that and tell you what resources you need to access it. Here, it is particularly vital that the resources are accurately priced and the cost of the process presented to the requester, especially if it involves taking reels of film to a telecine suite or old and potentially fragile videotape to a specialist transcription unit.
Some assets can never be digitized by their very nature. As part of its archive, a major production company will want to keep artifacts alongside the audio and video. The words of a script can be stored in file form, but the original shooting script with handwritten notes by the director is a valuable resource for future researchers.
Content security
The other widely appreciated benefit of the file-based environment is the ability to break out of linear processes and work collaboratively in parallel. There is no longer the structure by which one person performs a task on the master tape and then passes a new master onto the next process and so on. It has the potential of speeding up the workflow and should ensure that the content is technically safeguarded throughout.
The disadvantage is that, by making available the content to a number of people, the security of the intellectual property is decreased. That is particularly true for movies and big broadcast series, which may be shot in the UK; get CGI in the Philippines; be color-corrected in Santa Monica, California; and edited in Cape Town, South Africa.
Whether it is stored on the SAN in one facility or many, each is a risk. Even viewing material in the machine room has the potential for leaks: Who knows what has been plugged into the output on a jackfield? Every access must be noted and recorded.
Maintaining an audit trail for security is another process, which is best tracked along with all the processes and metadata by the asset management system. It will track access and digital signatures to build up a complete picture not just of what each file contains but where it has been and which individuals have viewed it or worked on it. Where files are transferred to other devices, such as an editing computer, it should also record that the files were wiped when the completed task was returned to the SAN.
Distribution
The final part of any process is delivery, and in the context of broadcast assets it is the ability to serve consumers with your content over multiple distribution platforms that raises the prospect of new sources of income. The challenge is that each will require a different combination of video and audio codecs and metadata structures. The BBC iPlayer has become massively popular, for example, but it is not a single portal but a number of delivery platforms, for instance sending Flash-wrapped content for the Web and H.264 for modern appliances such as the iPad.
Broadcasters, particularly in Europe, tend to outsource the delivery of their content to multiple platforms to specialist facilities. Each of these will be providing playout services for a large number of broadcast channels, some of which may be in direct competition. It is vital that the operator demonstrates that content is secure through an audit trail detailing who accessed it and why.
Content owners will distribute their material to aggregators such as the iTunes Store, which is strict in the way that metadata is presented. To be able to automatically generate successful packages for Apple, Amazon and others depends upon the right information being tracked through the asset management system. That data is then inserted in the correct format into the delivery files.
Distribution, then, is a matter of taking the content and processing it. In this case, it consists of upconverting or downconverting it, encoding or transcoding it, adding metadata in the correct format, and wrapping it for delivery. That process will take a number of resources. The asset management system should automate the process, calling in external resources as required.
But it will also allow the user to plan the workflow and estimate the cost, taking into account the technical and human resources required. Monetizing content in this way depends on knowing the cost of preparation and delivery, and therefore what overheads need to be cleared when setting pricing. That decision can only be taken in an enterprise view. Looking at a process in isolation will not reveal whether adding it to the system will reach the tipping point at which new capital investment is required.
In summary, to add maximum value to the business, content management should be extended into asset management, which takes an enterprise view of the content, its associated metadata including security, the processes that are applied to that content, and the resources available to carry out the processes. The result is a clear understanding of the costs and practicalities of doing business, as well as the management of all the processes. That, in turn, leads to a clear understanding of the return on investment in asset management itself.
Tony Taylor is chairman and chief executive officer at TransMedia Dynamics.
