The Insanity of the Monolithic Metaverse
Lessons from the Past and Building the Future
How can we help the next attempt at the Metaverse to be more successful? This article will present the idea that our attempts to directly build the large general-purpose virtual environments (“to build the Metaverse”) are, in itself, what have prevented a successful Metaverse from happening.
The Andromeda Blog warns us that virtual reality is doomed to repeat the failures of the past unless we recognize what those failures are, and start thinking in a new direction. They remind us that a popular definition of insanity is “Doing the same thing over and over again and expecting different results.” In the context of virtual reality, they’re right. We need to do something different than what has already been tried and failed.
What do people think is different this time around? “We have new technology!” “This time, we’re going to make virtual reality a platform!” “People are starting to take this seriously!” Those things are all important contributors, but are they at the heart of the problem? Only when we are able to recognize what we’re doing wrong are we able to figure out what needs to be different.
In previous articles, we’ve identified two easily overlooked but very substantial user needs which were neglected in previous implementations. First, there was a failure to maintain novelty (as the initial novelty decayed, large virtual worlds became boring). Second, there was a lack of utility (there was little real-world value which people obtained in virtual reality). A successful Metaverse has to continue to entertain its users. If and when that fails, it has to provide real-world value if it hopes to retain them.
What if there is another problem, more fundamental, that is baked right into the design?
Monolithic Worlds vs Modular Worlds
In this article, we’re going to explore the problem in different direction: monolithic worlds. This may not be a popular observation with some of the readers who find themselves invested in those designs. I understand; I happen to love a couple of them. But we need to illustrate a different approach.
Over the years, there have been many discussions between programmers about the trade-offs between monolithic and modular applications. Each have their own advantages. Many people seem to preach an approach that is somewhere in between the two extremes. I bring this up only because I wanted to use some of the descriptions in Mike Schinkel’s post not to talk directly about the code, but to describe where we’ve been with virtual worlds.
Look at the two groups below. Which one of Mike’s groups best describe the virtual worlds that you’re familiar with?
- Grand Visions
- [relatively] Significant Development Budgets
- Requirement to Accommodate Infinite Future Scope
- Minimum Utilization of External Components
- Software Comprised of Complex and Highly-Coupled Components
- Modest Visions
- [relatively] Little or No Development Budgets
- Moderate Consideration of Future Scope
- Cohesive and Focused Functionality
- Maximum Utilization of External Components
When I think of Second Life, World of Warcraft, Gary’s Mod, or perhaps even the multi-user dungeons of the 80s and 90s, I think of the first group. That group is monolithic.
The Monolithic Virtual World
The Andromeda Blog makes the distinction between a client-based virtual reality system and a browser-based VR system. For the purpose of this discussion, the underlying technology doesn’t matter so much. A monolithic virtual world envisions itself as an all-in-one solution that others will create their content inside of.
A monolithic virtual world generally begins with a moderately funded developer and a grand design. They create a general-purpose environment which runs within a tightly controlled specification. Yes, they might offer plug-ins or modules. They may offer you the ability to customize within their world, but the entirety of the world is under their control. You choose to build your own story inside of it. Not to the side. Not standing by itself.
When you build in someone else’s world, you are tied to their user base. You’re tied to their features. You’re tied to their upgrades and their innovation. You’re tied into their terms of service. You’re tied into their reputation. All of these can be positives and negatives. These alone are enough to keep developers at bay.
Does the engine for their virtual world lend itself to making a twitch-based shooter? A for-profit online movie theater? Secure chat? Real-world purchases? Telepresence? A hyper-realistic physics simulator? Low-latency communications? A monolithic virtual reality engine can’t be everything to everyone. It can’t be as good as a standalone application.
A previous article enumerated many of the choices in dealing with something as simple as load. Those are not fundamental decisions that you want someone else to make on your behalf. As an example, creating a twitch based first-person shooter inside of a virtual world simply doesn’t make any sense if that virtual world’s engine uses time dilation to handle load.
Are the faults of monolithic virtual worlds important? You better believe it. Innovation happens at the edges. Innovation is tied to novelty and utility. Novelty and utility are things that provide actual value to the end users. Trying to shoehorn many small developers into the same monolithic environment does not promote innovation.
When a well-funded company decides that they’re going to create “the Metaverse”, their design considerations will be shaped by the need for financial returns on a large scale. The natural tendency will be for them to go with a general-purpose monolithic approach. Do we have any reason to believe that “this time” they can use the same tried-and-failed approach and see different results? Does more money or newer technology cure the insanity?
A different approach is needed here.
The Modular Virtual World
If we see Second Life as an example of the monolithic approach, then what is an example of a modular approach? Something like Riftmax Theater would be part of a modular approach to a virtual world. Riftmax Theater stands alone and doesn’t require someone else’s virtual world to operate. It doesn’t try to be everything to everybody. It has a limited scope, and it concentrates on performing a core set of features very well.
The Opportunity to Do Something Different
Riftmax Theater succeeds, but it stands alone, unconnected to other worlds. How do we fit this and other well defined modules (with their unique code and focused advantages) into a larger universe? How does this form the Metaverse? This is the new problem that we need to solve.
While we recognize the value that monolithic designs provide, the design pendulum needs to swing back towards the middle. In a time of innovation and growth, we need the benefits of a modular virtual world. We need many developers working off of unique visions. We need competition.
Is there a way that we can build virtual worlds differently this time? Can we find a way to balance a general-purpose monolithic architecture with a specialized modular implementation? Can we unite stand-alone software? I think we can; it is just a matter of how. There are several ways to approach this, and an upcoming post may present one such solution.
- Augmented reality
- Data Collection
- Intellectual Property
- Science Fiction
- Second Life
- Virtual home