No metaverse in sight (& 2)
The idea of the countless “virtual worlds” that the metaverse will contain is, as we saw in our last entry, an extrapolation of the idea of “video game worlds,” not necessarily regarding their playful nature—since the metaverse is expected to be a place for almost any kind of activity, not just leisure—but in the sense that they will be artificially generated environments based on what we do in them. Why do I claim that it is so difficult for this to become a reality in the coming decades? The main reason for my scepticism (following Matthew Ball) is the immense technical complexity required to ensure that each and every one of those worlds can efficiently provide the three following things: (1) it works for any possible number of users, (2) it operates synchronously enough so that each user’s actions are instantly reflected in the interfaces of the other users involved, and (3) all the different worlds are fully interoperable, so that what happens in any one of them can have the necessary reflection in those other worlds where it is relevant.
Regarding the first point, of course there will be many worlds where the number of users won’t be a serious issue, as some may even have the advantage of not being open to everyone (they will be like a private club, so to speak). However, a metaverse worthy of the name will need to contain many open pathways for any user to navigate and, above all, will need to record what they have done there. This means that the amount of information that will need to be stored, just to even have the possibility of countless virtual worlds worthy of the name existing, will be several orders of magnitude greater than the amount of information expected to be gathered from the “real world” once the so-called Internet of Things became fully operational (that is, when every one of our real-world activities can be recorded in the cloud, from our heart rate during exercise to the contents of our smart fridge so that it can order new groceries from the supermarket). In other words, imagine an “Internet of Things”—which already requires massive storage and processing capacity—multiplied by the number of public worlds contained in the metaverse, and that will give us a minimum threshold for the amount of data storage required to implement a metaverse deserving its name.
Regarding the second point (synchronicity), most current video games are not truly synchronous. Instead, the server where they are hosted sends and receives data packets to and from our device in the smallest possible sizes to make the gaming experience feel like it flows naturally. Since the informational content of these “worlds” is actually very limited compared to the overabundance of the real world (for example, the driver of the car we are chasing in a video game won’t remember the names of their grandparents if we manage to stop them and ask, and the car’s body won’t tell us the metals it’s made from if we subject it to a physico-chemical analysis), the trick usually works in video games, giving us the impression that “everything has always been there” when we need to interact with it. However, if the actions of thousands or millions of people who may be participating in the metaverse worlds are required to be immediately reflected in the interfaces of all others who may interact with them, and this happens not based on a tiny, closed catalogue of possible responses each person can give, but with the spontaneity, variability, and unpredictability of real-world interactions, the capacity to store, transmit, and process all that data at high speed seems far from feasible in the medium term.
Finally, the third difficulty (the lack of interoperability) might seem less significant. After all, a similar problem existed in the early days of the global network (the World Wide Web or www) until protocols were unified to allow our browsers to access any internet page (protocols like TCP/IP, HTML, or URL). But in the case of the metaverse, the problem is multiplied by the complexity of the information that needs to be handled in each “world” (for example, when dealing with three-dimensional environments rather than just text and hyperlinks). Interoperability between the various virtual worlds is necessary so that these are connected by more than just the fact that the same person visits different worlds at different times. It’s about ensuring that certain actions in one of these worlds (e.g., buying or selling a product) can have consequences in others (in this case, being able to use in other worlds that product or the money earned from its sale). If the metaverse aims to be an economic reality and not just a collection of independent games, most of its “worlds” would need to allow for “imports” and “exports” between them, i.e., the possibility of using in one world what was produced or done in another. If this kind of interaction isn’t possible—and the complexity of the challenge leads to pessimism on this issue—the metaverse would be reduced to a series of isolated compartments, which would cause it to lose one of its main supposed advantages: the possibility that interactions between various worlds could create unexpected innovations.
To conclude, I want to add a fourth reason why I believe the metaverse may not turn out to be as spectacular a phenomenon as its proponents promise to us: I highly doubt that the types of interactions that can take place within it will be attractive enough to a sufficiently large number of people to justify a massive investment in this technology. We can divide the activities that will occur in the metaverse into four main categories.
First, there are what we might call extended video games, meaning simply more sophisticated versions of current video games. This will surely continue to develop, but it won’t constitute “a new reality” much more than current video games already do. A second category would include virtual spectacles, which I believe (to the extent that they aren’t video games and don’t allow for much audience interaction) would merely be slightly more dazzling than current films, operas, or concerts. This means that the creation of “new universes” as compared to current technology would be in the same order of magnitude as television did compared to cinema, or cinema compared to theatre, but not much more.
Third, there would be virtual services—things that now require visiting a store or an appointment (medical, legal, technical…)—but these would be little more than a small improvement over current online shopping and phone consultations.
Lastly, the fourth category would encompass what would be more characteristic of a true metaverse worthy of the name: spaces for interaction with others for all kinds of activities, not necessarily “programmed” (in the way a show, a game, or a medical appointment is). I’m referring to something like transferring activities such as travel, politics, and much of our work (such as education) into the metaverse. I doubt this last category will be particularly appealing to most people, as physical interaction will still be desirable in most cases. The dream that these types of activities in the metaverse could one day compete with the real world seems as naïve to me as when, in the early days of the internet, we were told that we would no longer be limited to reading the newspapers sold at our neighbourhood kiosk, but that we could have breakfast while reading newspapers from anywhere in the world (I even spent some time browsing African newspapers). Well, a quarter of a century later, we have that possibility more accessible than ever, yet the vast majority of us remain far more interested in national politics than in what happens thousands of miles away, and we’re usually satisfied with the version of international news provided by the newspapers around us, with the occasional click on a foreign source. In the case of the metaverse, I predict that most of the so-called “alternative realities” being promised will also end up boring us after a few weeks.
References
Ball, Matthew (2022), The metaverse, Norton & Co.