eliminate them because the content ultimately has to reach the viewer.”
However, the panel stressed that any transition would take many years.
The panel agreed that piracy has become increasingly sophisticated, with illegal streaming platforms now offering polished interfaces that rival legitimate services. Strong said DRM remained essential despite its limitations. “It’s fundamentally an arms race. You’re trying to make it harder for hackers to steal your content, but it’s almost impossible to stop them completely because, ultimately, legitimate viewers have to be able to watch it.” Looking ahead, the panel predicted that consumers will eventually stop thinking about streaming as a separate technology. Nock suggested that within a few years, internet delivery will simply become “television”, while Siepe anticipated increasingly personalised viewing experiences driven by artificial intelligence. Strong concluded that although new technologies such as Media over QUIC may transform the technical infrastructure behind the scenes, viewers are likely to value something much simpler. “If there was a single thing my wife would like when she’s watching Netflix, it’s that she doesn’t have to spend 15 minutes trying to find something to watch. That’s where I think AI could make the biggest difference to the viewing experience.”
Nock pointed out that the industry had already spent around 15 years refining HTTP adaptive streaming. “We’ve had a good run with HTTP adaptive streaming. We’ve improved it over many years and we’ve built a huge installed base of devices that support it. The next transition is going to take an interesting amount of time because we’re now talking about replacing technology that’s already in millions of homes.” Strong predicted that MoQ would initially be adopted for contribution links before gradually reaching consumer services, while older technologies would continue to coexist for many years. Quality of Experience (QoE) was another major focus. Siepe highlighted the growing role of Common Media Client Data (CMCD), which allows streaming players to send performance information back to content delivery networks. Rather than simply measuring whether a stream is working, CMCD provides detailed information about buffering, bitrate switching and playback performance. “What broadcasters really want is observability across the whole delivery chain. They want to know if there’s one specific device that’s failing, or one ISP that’s causing problems, before viewers start complaining. That’s where this kind of telemetry becomes extremely valuable.” He said AI was increasingly being used to analyse enormous volumes of operational data automatically. “We try to code a video streaming engineer that looks at an issue and tries to understand what the root cause could be. The aim is to reduce the engineering effort needed to monitor these systems while spotting problems much earlier.” Security and piracy also featured prominently. Nock described streaming security as an ongoing battle rather than something that could ever be completely solved. “I always describe it as a leaky bucket. The bucket still has to work as a bucket because people have to be able to watch the content. You can make the holes smaller, but you can never completely
Made with FlippingBook - Online magazine maker