Where will users be located when they are participating in metaverses? In scenarios currently being discussed, actual users may be inside physical buildings, specialized facilities, outdoors, vehicles, and more.
There may be some entirely local metaverses. But even for local events, there will be a pull to go global. Let’s take “Hardly Strictly Blue Grass” (HSBG) as an example. Today, HSBG is a three-day series of free concerts held annually in San Francisco’s Golden Gate Park. During the pandemic, part of it was held online. Musicians still traveled to a stage in San Francisco (at a secret location), but their performances were available on the web and watched by people all over the world. The pandemic produced this as an emergency response without an existing metaverse infrastructure to build on.
Let’s imagine that HSBG decides to go to a metaverse. Now, the global audience can interact with each other. The musicians don’t have to travel—in fact, the musicians don’t even have to be co-located. There may be vendor booths that users can enter to purchase music, swag, and more. At some music festivals, there are food vendors. You could imagine a food delivery service partnering with the festival to deliver food to audience members. HSBG is free and held during daylight hours. In the evening, many of the musicians in town for HSBG play in clubs and halls—these venues could also be part of the metaverse. This is just an illustration, but when creative people fully engage with metaverses, the results will probably be well beyond what can be imagined now.
Many special events like HSBG, plus ongoing virtual worlds, will dramatically increase communication volumes. But that is not too hard to handle with today’s technology. The greatest challenge will be to meet the latency requirements. For example, such a metaverse HSBG could be hybrid—that is, both in person in Golden Gate Park in San Francisco and virtual with attendees and performers from around the world. Delivering the performances from the physical stages could be handled similarly to streaming movies or Zoom. However, virtual participation, particularly concerning performances that combine both physically present and virtually distanced performers, will create demands for very low latency. These demands for very low latency will be difficult for today’s packet-switched networks to meet.
Packet switching played a key role in our transition from analog to digital communications and is a fundamental part of our digital infrastructure today. The most common form of packet switching in use today is built on TCP/IP. There are many benefits enabled by packet switching. But there are costs as well—and one of the cost factors involves latency.
To understand latency, it is helpful to use an oversimplified description of how a packet network works. This description is designed to make sense to people not familiar with the details of packet networking and to be consistent with the basic understanding that those with detailed technical knowledge have. In a packet-switched system, every node in the network has to process the packet. This means, at a minimum, the node has to read the top-level packet address and make a decision about what next node to send it to. Nodes may also read some, or all, of the progression of lower-level