Signals go over redundant paths through the Telstra network. Latency from cameras in Perth to switching in Sydney or Melbourne – 4000km away – is said to be 27ms, which seems pretty good to me. “We’re live tonight – but from 400km away. The only thing that is in Perth are the cameras,” said Mark Segar, NEP Australia director of technology.
That is the primary hook for remote production: people don’t need to go to the ground. No-one really relishes the prospect of, say, a February evening at St James’ Park in Newcastle in the snow. Nor, to be honest, is there much to be said for a May test match at Chester-le-Street. Keeping everybody back at base, where the chairs are adjusted to perfection and everyone knows where the nearest/best coffee bars are, seems like a perfect solution.
Except, of course, you don’t keep everyone back at base. You have to send riggers to get the equipment in the right places, and camera operators to point them at the action. You have to send engineers to connect the broadcast hardware to the data encapsulators and onto the internet. You actually have to send an outside broadcast truck to the venue, even if it the equipment inside is a bit different.
You probably have to send the commentators and reporters. Commentating off the tube is never the same thing, and anyone who has tried pitch-side reporting knows that the only way to do it is to grab someone and stand them in front of a camera.
Over breakfast this morning I was discussing my ideas for this article with my wife, who understands a little bit about broadcasting. “What about when something unexpected happens?” she asked me. “A drone lands on the pitch, or something?”. I hope never to see a Hillsborough or a Bradford City disaster again in my lifetime, but sometimes things happen that are completely unplanned but you have to cover.
So do you have to have producers on site, to plan interviews and to be the resource to lead coverage of the unexpected? Should the worst happen, who makes the decision to lock the cameras off and head for safety? Can you reliably make the fine distinction between an instruction to run or to pick up your camera and run from 4000km away? None of this is quite as simple as it seems.
Related to remote production, I predict that IBC will see people talking about 5G as it becomes a reality. The BT/EE demonstration broadcast from Wembley Stadium over 5G is an entrant to this year’s Awards, and it is another important technical step in remote production.
One fascinating entry which all the judges loved came from Romania. The story is that a 10 year old guitar prodigy was invited to play on stage with his favourite band, but was struck by stage fright. The solution was to put him in a room the other side of town, capture a 3D image of him playing, send it over a 5G link, and show him on stage as a “hologram” (actually a good Pepper’s ghost).
The video clips suggest it was very successful: again, minimising latency is vital if you are going to meet even heavy metal’s timing standards.
It’s an interesting and unexpected use of 5G, and I am sure that we will see many more things crop up that we cannot conceive today. It is going to be a critical plank in the internet of things, which is another hot topic.
According to top forecasters IDC, by 2025 the world will have 41.6 billion “things” connected to the internet, generating 79.4 zettabytes of data. Yes, I had to look up precisely how many noughts a zettabyte has: it is 1,000,000,000,000,000,000,000 bytes, or a million petabytes, or a trillion gigabytes. Or 1,500,000,000 CD-Roms.
According to Cisco – and they would probably know – the whole world’s collective internet use in 2016 was one zettabyte or so. Less than a decade later, machines will be creating 70 times as much data on their own, let alone what we humble humans will be doing.
One of the big data uses will be driven by artificial intelligence, which is my final pick for top topics at IBC this year. NAB saw a lot of stand graphics proclaiming artificial intelligence with very little evidence of anything beyond the ability to make decisions based on business rules. I am confident that, both in the conference and on the show floor, there will be some sensible applications.
There will also, no doubt, be discussions on the trend of deepfakes: the use of AI-boosted software to create videos of things that did not happen. Given the calibre of leader we currently have in the West we have more than enough video clips of politicians saying stupid things.
But the fact that it is perfectly possible, to pluck an idea at random, to have Albert Einstein discuss Game of Thrones, allows the unscrupulous to throw around charges of “fake news” even when the material is genuine. That is a worry for the future.
But perhaps not as worrying as a report on the BBC News website on the day I am writing this. The first paragraph I quote verbatim: “An app that claimed to be able to digitally remove the clothes from pictures of women to create fake nudes has been taken offline by its creators.”
Yes: for $50 you could buy Deepnude, an AI app (Windows or Linux – us Mac users are way too principled) which uses neural networks to plot where the clothes are, estimate the body shape beneath, and paint it in matching skin tones and the original photograph’s lighting.
The developers are quoted as saying “the probability that people will misuse it is too high”. Yes, I think that probability cannot be underestimated.
That’s what I am looking forward to at IBC (apart from the fake nudes, of course). What is on your list? What do you think real people will be looking to debate and, more important, spend money on? Let us know what you are looking for – and if you find it.
Oh, and just because I have mentioned some of the shortlist for the IBC Innovation Awards and not others, that in no way should be interpreted as an attempt to influence the final result!