Playout & Transmission Global Viewpoint – October 2021

Broadcast engineers are known to be hoarders. In every facility there is a room that’s full of bits of obsoleted equipment that might just come in “useful one day”. Shelves of circuit boards, cables, and assemblies are prevalent and most of them will never see the light of day.

There’s an interesting analogy between agile and the modern minimalism movements. The internet is full of advice on how to declutter, own only ten items of clothing, and scan your book collection, all with the aim of freeing your mind. The implication being that owning stuff will stop you thinking straight or somehow question your powers of judgement. Strangely, I think there may be something in this.

Rip-and-replace is the idea of deleting virtual machines when they’re no longer needed, not archiving, or storing them, but throwing them away. Even fixing them can be frowned upon.

Intrinsic in every engineer’s DNA is the need to take things apart to see how they work, whether it’s their car, washing machine, or camera, there is an inherent need to get the screwdrivers out and have a look inside.

So, what about the cloud? We can’t see it, touch it, or smell it. If we’re lucky and have a private datacenter then it may be possible to take the lid off a server and have a look inside. However, the abundance of SLAs and manufacturers warrantees soon put stop to this idea.

Although software tools have taken over the soldering iron and tweaker, the great news is that we can still build our own tools. Any engineer working in a modern broadcast facility will need to understand how software logging and monitoring works. They will need to write Python scripts and C programs in Bash shells to enable them to gather and analyze Gigabytes of logging data.

In essence, the same “take things apart” mindset still exists in the cloud, it’s just that we can’t physically hold the screwdriver. And being proficient in writing code is a major requirement for the 21^st century broadcast engineer. The new breed of microservices have APIs that allow us to mine a massive amount of information so we can understand how systems are performing. We can see where the bottle necks exist and tweak the workflows accordingly.

In the cloud we still use the same mindset we used to build custom tools, and operate AVOs, waveform monitors and oscilloscopes, it’s just that the end product is something we can’t physically touch. Operating the cloud demands the ability to code which in turn releases broadcast engineers from the physical constraints of hardware.

In a strange way, throwing away the shelves of obsoleted kit, fix-it boxes, and custom tools does release our minds. The rip-and-replace mindset demands a new discipline, one that frees us from old static ways of thinking in favor of the new dynamic, scalable and flexible approach. I accept building your own software tools is a leap of faith, but it is one that will continue to deliver.