SPECIAL REPORT: Implementing new workflow principles

If you ask a dozen vendors what media workflow means to them, you'll get two dozen answers. With that said, let's develop one and stick with it throughout this article. This definition can be applied to a wide variety of needs yet covers the entire media lifecycle:

Media workflow is the movement and transformation of content across the three fundamental activities of the broadcast business: acquisition, management and distribution. Movement of content through each of the high-level activities is managed by a series of discrete operations called workflows.

With that stated, the strategic objective of successful media workflow management should be to add value or remove cost at each step of the process, not just the opposite, which can often be the case.

Doing more with less

Broadcast engineers are faced with the challenge of reducing the number of discrete steps required to move content through its lifecycle while assuring that critical business processes remain unaffected. The underlying, strategic vision should be to design a technical and operational system that completely automates the movement of content from beginning to end while simultaneously adding value and removing costs at each step of this process.

An ideal workflow system will eliminate ad hoc exceptions and follow role-based business rules. Such a workflow will seamlessly integrate and automate as far as possible the following steps: acquisition, ingest, metadata acquisition, storage, asset management, search, retrieval, desktop cuts-only editing, trans-coding, scheduling, automation, playout, billing, accounting and archival management.

To accomplish this, secure, scalable systems must be designed and built incrementally in ways that preserve legacy infrastructure investments. Also, these systems must allow media assets and asset-related data to be available to enterprise information systems that manage financial, customer, sales and rights information and reporting. In other words, Mr. Engineer, this solution is a whole lot bigger than you may have thought.

Analog today

Let's examine an example of an analog linear workflow. (See Figure 1 on page 42.) Consider a large media enterprise with the following attributes:

  • The need to feed multiple output channels.
  • The need to develop fresh, updated programming on a minute-by-minute basis.
  • Incoming feeds from multiple sources, many arriving at once.
  • A large and growing analog tape library.
  • A laborious process of getting archived tapes.
  • A chaotic production process.
  • A linear editing process.
  • Tape-based analog playout.

Consider the steps required to repackage a piece of content that resides in the tape archive. An analysis of the steps that must be completed to simply identify what tape is needed, get it out of the library and get it on the air requires at least 14 discrete steps and decision points. Four employees may be involved just to find the tape (if it can be found at all), get it back to the production room and put it in the hands of a producer. Even worse, it might take several hours to go through this process. Multiply these steps for every tape request, and you can begin to see the inefficiency of this system.

Even if all the tapes are found, they must be reviewed and ingested to the production system before they can be re-edited, after which they have to be played out to tape again and delivered to the NOC or MCR for broadcast.

This disarray is the result of a process that is broken from the very beginning. Consider how the tape originated in the first place. The feed is first recorded to tape. The metadata probably consists program name, date, source and length. That information, if it's recorded at all, is logged in real time by hand, scribbled on a piece of paper and stuffed into the tape box. Later, the tape is sent to the library where the metadata, if legible, might be entered into a home-brew asset management system database and finally the tape shelved according to who knows what organizational system. If you need that tape in the interval between taping and shelving, you're probably out of luck.

A digital workflow

Let's look at this same set of tasks at a facility where the broadcaster understands the value in a digital workflow. Further, let's assume the station recognizes the benefits of being able to easily reuse assets to expand market share or develop new products. New products could include developing a third or fourth newscast each day or content for a cable system. Or perhaps the station is part of an LMA or wants to begin centralcasting.

Obviously, any engineer worth his salt would be smart enough to know that only a completely integrated digital-based system with automated workflows can meet this mandate. Also, this savvy engineer will recognize that no single vendor's software product can come anywhere near addressing the massive complexities in designing and building such a sophisticated system.

As a result of our brilliant engineer's intensive analytical and development efforts, the system he's looking for will completely automate the acquisition, capture, management and storage of new content. Let's examine what this new automated workflow management system looks like and some of the benefits it can offer. (See Figure 2 on page 42.)

Digital workflow

First of all, scheduled feed requirements reside in a database that triggers multiple events: orienting the satellite dish, selecting the proper transponder and channel, routing the feed to the station's ingest server and finally begining recording the feed. However, the process isn't complete.

As soon as the first frame of the feed has been captured, the system makes it available to everyone throughout the enterprise, depending on levels of permission and security. Basic metadata is captured along with the feed and can be used for immediate search and retrieval on standard desktop computers. This can be done by proxy or full-bandwidth images if the LAN has sufficient bandwidth.

Producers can identify and retrieve current or legacy content (if it has been appropriately ingested into the system) and create programs on the fly. If craft editing is required, an EDL is sent to a craft edit suite and the content is pulled into the NLE, where graphics, music, narration and effects can be added. Conforming and rendering is done in real-time, and the content is instantly forwarded to the play-to-air server and immediately available for broadcast.

Workflow benefits

Now this may sound like utopia, but it is possible to come quite close with today's technology. Let's examine some of the potential benefits.

The time required to locate and retrieve content online goes from hours to seconds. No more looking for tape boxes. Requests are now filled from an on-screen database. Lost content happens less often. Once the material is stored and tagged, it's virtually impossible to lose the material. Compare that with trying to find that partially labeled box containing the tape of last week's fire.

What about creative flexibility? Before, editors were limited by their access to the content and crushed by deadlines. Now, content is always accessible and instantly deliverable. Also, editors can share content. No longer do they have to wait for a dub to begin work.

Staff changes required? You and your staff get to keep your jobs. Responsibilities increase, and people focus on what they do best — creating programming. Finally, because such a system can be built with nonproprietary hardware and standard operating systems, it's easy to scale, maintain and integrate with an existing IT and broadcast infrastructure.

Scale to fit

Such an enterprise-wide workflow isn't cheap. For global media enterprises, those costs are justified by the potential to reduce operating expenses and realize new revenue streams from a system that integrates broadcast, cable, new media, and handheld device distribution. For smaller broadcasters, you may not need to build your own solution at all. It's often possible to access these services on an outsourced or shared basis.

Consider the options

Suppose you decide to build an in-house system. What are the options? The first is to build your own solution. There is a natural temptation for IT and broadcast engineers to want to demonstrate their skills and expertise to senior management. And, engineers often like having control over the solution. But a word of caution: Such systems are complex to design and challenging to build. When there is no out-of-the-box solution, this means that no single hardware or software vendor can provide more than a portion of the system.

This will require the staff to integrate all pieces of the puzzle. Also expecting multiple equipment vendors to integrate with each other will greatly magnify the chance for failure, especially where software interoperability is critical to success. Home-brew systems are rarely successful, and they can easily create more problems than they solve. If that happens, the finger of blame will point only to you.

Another possible solution is to outsource to an IT consultant. Traditional IT consulting firms are great for enterprise systems, e-mail and desktop support. However, few of them have the specific workflow expertise required to successfully design and build systems that automate the complex interdependencies in a broadcast station.

A third option is to outsource the project to a traditional broadcast integrator. This option will give you access to broadcast infrastructure design and build experience and access to broadcast hardware sourcing. However, building the needed IT-based systems requires specific IT and programming expertise, skills that may not be available from your broadcast integrator.

A fourth — and, in my opinion, the best — option is to partner with a firm that can staff both sides of the challenge with broadcast expertise and IT knowledge. Look for demonstrated skills in designing and deploying complex, enterprise-wide broadcast-IT integrated systems. You don't want to be anyone's guinea pig when it comes to your IT broadcast project.

Fortunately, there are firms that can meet these requirements. But, before you make a decision, do some homework. Look at the company's track record. Call its customers and then visit those sites. Look at both newly-completed sites and those with many months of operation under their belts. There will always be start-up issues in complex systems. What you want to know is whether the vendor took care of them and whether the system works reliably over time.

Don't be lulled into thinking that building an enterprise-wide digital workflow is merely connecting the dots. It isn't. Broadcasting is one of the more complex media workflows out there, and you shouldn't be timid in saying that you need help in finding and building the best solution.

After all, when the project is completed, you're going to get the recognition for the project anyway. Wouldn't you rather that it be for a job well done?

David Rosenberg is the media consulting practice manager for Siemens Business Services in North America.