On Automation.

Picture the scene. You’re sitting there watching television, and something bad happens. Maybe it’s a voiceover in the wrong place, over the last scene of the programme instead of the end credits. Maybe the channel goes to an ad break in the middle of a scene. Maybe the first part of a programme keeps repeating over and over again. Maybe the aspect ratio has all gone to shit. Maybe it’s a full-blown breakdown, badly-dealt with and with no apology.

And across the internet, the familiar lament goes: “Tch, automation, eh?”

Except: it isn’t. Automation doesn’t really have anything to do with it at all. And I’m going to do my best to convince you. So what does cause complete inanities to go to air?

1) How many channels is the operator in charge of?

Oh, yes, channels do indeed have operators.¹ It is very, very rare that a channel will routinely be left with nobody looking after it, even on the ultra-dirt-cheap ones. But are we talking about single channel playout – one TX operator in charge of one channel? Or are we talking about multi-channel playout – one TX operator in charge of, say eight channels?

Having done both², I would argue that despite sharing a title, they feel like almost two entirely different jobs. With single channel, the op has more of everything: more time to check and prepare the schedules, and more time to monitor the output – and notice when things go wrong. Multi-channel will inevitably have less checking – and monitoring eight channels at once is very much not the same thing as carefully keeping an eye on a single channel.

The key thing here: both single channel and multi-channel setups both use automation. They both use playlists which automatically play material from a server – with varying amounts of manual intervention from the ops, of course. But it’s not the automation which is important here. It’s the prepping which stops errors getting to air, and it’s the monitoring which means errors are spotted, dealt with, and hopefully won’t make it to air again. All of which is far more comprehensively done in a single channel environment. That’s the reason why things this like this rarely happen on the major UK networks. Automation has precisely nothing to do with it.

Of course, it’s worth noting that the introduction of automation is what made multi-channel playout possible in the first place. Moreover, automation has caused the number of people required in even a single-channel environment to shrink. But that doesn’t mean that it’s automation itself that should be blamed: it’s how you use it which counts. Whether you’re firing off a VTR manually or playing automatically from server with your beady eye on the output doesn’t make a scrap of difference.

2) Has the channel got a live announcer?

Yes, live announcers are alive and well in 2015 – on the major channels, at least. And a skilled live announcer can make all the difference between a professionally-handled breakdown, and the channel looking a complete mess. Admittedly, pre-recorded announcements can be deployed to good effect by a skilled operator, but it’ll never match what you can do with someone sitting in a booth, ready to leap into action at a moment’s notice.

It’s never breakdowns themselves which really annoy me as a viewer. It’s when they steadfastly refuse to acknowledge the problem on-air that I feel like a channel is insulting my intelligence. Indeed, a breakdown which is well dealt with increases my respect for a channel, not diminishes it.³

Of course, it’s not just breakdowns where live announcers can be useful. If you have a live show that has overrun, or a late schedule change, they are absolutely invaluable. And the odd topical reference in an announcement can make a channel feel alive, not just a box in the corner running a playlist.

So, yes: live announcers. Nothing to do with automation. Next!

3) How easy is it to switch to backup systems, or bypass the automation entirely?

No matter where you are or what channels you’re working on, any TX op who has been in the job for a reasonable length of time will eventually have the automation fall over on them.⁴ How stable the automation is in the first place is a consideration of course, although you’ll rarely find an op who is truly happy with everything. But when things go wrong, you need an easy escape route. And preferably an escape route which can make things look as smooth as possible to the viewer in the process.

How easy is it to switch to a backup? How easy is it to bypass the automation entirely and route a source direct to air? Is there a tape backup available, or a reserve line for an incoming live programme if the main one goes down? And, for that matter, how experienced is the operator in dealing with these things, and how experienced are the engineers in fixing the initial problem?

All of this has nothing to do with the concept of automation per se: it’s about staffing, training, budgetary concerns, and technical aspects of the suite.

I could go on, but I think you get the idea. To be fair, in many ways it’s easy to see where the use of automation as a bogeyman has come from. Yes, television gets worse the more the human element is removed. It’s not a huge leap from that to an all-encompassing “blame automation”. But automation still has a human element behind it. It’s how much the human element gets deployed which is key. When you’re sitting there and watching BBC One, BBC Two, ITV, Channel 4, or Channel 5, you’re watching a channel run via automation. But behind-the-scenes, there is one person, in charge of one channel, with live continuity for the vast majority of the time⁵, with plenty of manual intervention.

On the other hand, there are the channels further up in the EPG. One person looking after eight channels, far less prep time, far less able to closely monitor the output, no live announcer, with fewer facilities to fix things if they go wrong, and with very little power to suggest any amendments to what goes out anyway. Those channels are also running under automation… but under very, very different circumstances.

Automation is never the problem. It’s whether the staffing and processes behind it are sensible which is the key thing. Which may, ultimately, be a boring answer… but it doesn’t stop it being true.

1. For the sake of convenience, I’m going to use “TX op” or “operator” to cover a number of different names for roughly the same job. “TX controller”, “Playout Director”, and “Network Director” are also used. Best check which one to use in a given situation when talking to someone who works in the industry, or you might get a punch in the gob. ↩

2. I’ve also had the odd situation being in charge of two channels – a halfway house between single channel and multi-channel. To be honest, all that feels like is that you’re supposed to deliver single channel performance despite the fact you have two, which is a tad annoying. ↩

3. For the record, I think a live announcer should sound slightly surprised when they apologise for a breakdown; you run the risk of the breakdown feeling weirdly planned if they don’t. The announcer should reflect how the viewer felt when their screen went black. Possibly with less swearing, though. ↩

4. In my experience it’s often issues with the mixer, not the automation per se, but this is a distinction you don’t need to worry about unless you’re being paid to worry about it. ↩

5. Channel 5 only has live continuity between 5pm – midnight. ↩

Read more about...

best of