Software Infrastructure Global Viewpoint – July 2022

Generalist Or Specialist?

Television has always been a niche industry, and broadcast engineers have been by default, specialists in their field. But is it now time to start thinking more in generalist terms?

The reason I ask this is in part due to the massive advance in technology our industry has witnessed over the past five to ten years. We’ve certainly seen and embraced our fair share of change, from analog to digital to HD, 4K and now IP. At the time the change from analog to digital may have seemed like it was happening at break-neck speed, but compared to the developments we’re seeing with IP, it was relatively slow.

Not only is IP paving the way for the future of television, but we are also making it work in ways it was never designed to operate. By design, IP is an asynchronous system that excels in sending short bursts of data between hosts through networks that effectively manage themselves (except for SDN). However, both video and audio used in broadcast television fundamentally rely on synchronicity between the sender (camera or microphone) and receiver (monitor or loudspeaker).

This all leads to a system that has become highly specialized but relies on COTS and generic system design. In essence, anybody should be able to make a studio-based IP network that reliably distributes video and audio with minimal latency. But anybody who has tried this will tell you how difficult it really is, mainly due to timing, latency constraints, and network reliability.

Surely then the modern broadcast engineer must be a specialist? It would appear so! However, can one person specialize at everything? For example, the same broadcast engineer may well need to write Python control and scheduling programs that interface with microservice APIs or configure Wireshark to capture IP streams and analyze them to determine the source of an intermittent network congestion issue or find out why a Dante microphone isn’t appearing at the sound console, and so it goes on.

The frustrating part for broadcast engineers working in modern facilities is that they have a natural intuition for developing any of the skill sets needed. However, what they lack is time as there is only so much one person can learn and do in their lives. The challenges of fault-finding IP networks are immense, especially when we start considering PTP reliant systems. We simply cannot treat network systems in isolation as the cause-and-effect influence continues with ever increasing effect.

Broadcast television and all the knowledge needed to make it work requires a very specific discipline in its own right. Added to that the complexity of transporting synchronous signals over asynchronous networks along with fifty years of IT history and established working practices, creates a skill set that no one person ever hope to achieve. It may be that broadcast engineers should be content with being generalists, but the generalist bar is very high, almost to the point where we are specialists!