In the past year (2021) especially the whole tech bubble went on and on about the metaverse. But I think it is premature to announce its conception. I have a very clear vision of what the metaverse ultimately is and thus of the prerequisites needed to actually make it happen.
Obviously there are a lot of definitions of the metaverse out there, but I think most of them are missing the point of what will be the next digital revolution. So here we go.
The next digital revolution.
Compared to the last digital revolution, this one will be even more drastic and it will touch even more aspects of our daily life. Not only do we need to change our understanding of how we deal with data as the world wide web did once. In addition to that we need to change our understanding of computing as well.
The Metaverse will drastically change what privacy and property mean for hardware, software and data.
Let’s start at the end here. The Metaverse for me is a “compute anywhere” scenario. With the most recent (www) digital revolution, we were suddenly able to access the complete knowledge (Data) contained in the internet at the press of a button. With the next one, we will be able to harness all computational power of humanity at any time.
This will not only encompass the actual processing resources, but also the algorithms that we as humans have created. This means, that fidelity, human and contextual awareness, uniqueness, personalisation and value of the output of every computational effort and every human-machine interaction will be on a whole different level.
It will enable an almost (not down to the molecule) complete digital twin of reality where every aspect of reality is linked to every other one. But have no doubt: It will happen in the pursuit of making (more) money.
The concrete future.
You will be able to answer any question regarding any known detail physical aspect of our world (known meaning, there is a feasible explaination): When was this building constructed? Where did the bricks come from? Which companies were involved in the construction? Which brand and product type is the tap in the bath room? How much would a new one cost?
You will be able to create, change, reproduce and remix every type of data at any time everywhere: How long is this desk? Would this plant fit and thrive in my home next to the couch? What would it look like standing there? Is there another item that would fit better especially with my taste and gardening skills?
You will be able to synthesize and simulate any situation: You can capture and relive a moment later, or create a scenario from scratch: What if I said something different? What if we moved on and did not stop here? Am I a good public speaker? Was I convincing the others?
I was specifically focussing on the value this tech will provide the end-user in the real world, because I think this is where this vision drastically differs from what is commonly portrayed as the metaverse. It will especially not require you to buy a specific device such as a AR/VR headset to profit from it. Instead you will have access to every bit of data and every processing unit at the same time. There will be few device dependant applications for enthusiasts, creators and/or developers. The rest of us will be able to get access using common, existing tools (eg. Smartphones, Laptops, Smart TVs). Displays and speakers will be accessible from any device no matter if you own the actual device or not.
Why this is not happening (yet).
While I could go on and on about the possiblities of the Metaverse, I am fully aware, that this is scratching the surface of what will be. Instead it might be more insightful to ask what we need in order to actually come closer to it.
For this version of the Metaverse to exist, I image five pillars, that are drastically different from today.
- Unlimited storage and the ethics of digital legacy and self-destructing data
- Time efficient creation and scanning processes
- AI needs to be ubiquitous to clean up, remix and use data anywhere.
- Changed understanding of Interlectual Property
- Shared computational resources
Unlimited storage and
the ethics of digital legacy
We need a way to store unlimited amounts of data and (the and is important here) know how and when to destroy them. We need to evolve our ethics around data and legacy as a digital society. Currently data ownership and privacy are a huge issue as it is handled by huge corporations acting effectively as governments in their own right. An alternative solution would be to decentralize data and provide very granular access to the encryption of it for the owners to control (very tedious) or give up ownership entirely and instead opt for a vast data lake for all human data. Here the value of the data comes not from the data itself, but from its connection to other data. And the connection can be either public (linking to other data of similar caliber) or private (linking to a specific person). Data in essence needs to be able to self-destruct once it is no longer relevant (eg. no connection can be made to another datapoint anymore or if it simply expires).
Time efficient creation and scanning processes
The speed at which data (especially 3D-Data) is created especially using AI, photogrammetry and similar non traditional techniques is increasing. However we are still missing a consumer device that makes the scanning process a commodity. As the smartphone did with taking a photo, it should be as easy as point and tap. In addition, we need tools/ways to enhance this data later (say if someone scans the same object in better conditions or data can be augmented better because a new connection with other data was made). To make sure it can be augmented by someone else, we need an agreed-upon, open data format for this, which we are lacking right now. Only then will we be able to effectively create a digital twin of our physical world.
AI needs to be ubiquitous
As has become apparent in the past few years, AI will define a new set of tools, we can use to create, access and manipulate data. Current AI solutions however rely on closed-source implementations run by specific companies working on data-sets, that are also not open. It is my strong belief, that by keeping these solutions closed, we are denied access to the single most important turning point in computing history. Only by opening up the AIs can we begin to build trust as a society and yield the innovation we as a society are capable of.
That being said, at some point in the near future we will see AI enabled tools for every digital process you can imagine. You will be able to mix, enhance and transform every possible data set using AI.
Changed understanding of Interlectual Property
We can already see a slight drift in how IP is managed today. Using IP in other third party worlds is already happening. This marks only the beginning though: Today you can ask an AI-prompt what a Super-Mario Mug would look like or how Jurrassic Park in Minecraft would be built as a world. And we will see more and more of that. In the past, the ways IP was used in third-party media was limited to the creative understanding of the people involved in the making of said media and the original IP. We will however see a shift where every third party IP can be used in any media for its own good. And the owner of the property will be able to benefit from it nonetheless. So it is not IP that is going away (or making money with it) – It is rather the way it is limited and controlled that will change.
Shared computational resources
Today access to computational power is still paid for and you can frequently read news about, who is building the next super computer. However, we will truly unlock the next big digital revolution, when we start sharing computational resources. We have to understand the resources we have on this planet as shared in general to be able to maneuver the climate crisis. The same is true for computational power: if we want to unlock the next level of our digital age, we have to give up ownership on computational resources, and build the tools to access all computational power everywhere with the press of a button. This would reduce complexity for edge devices, increase efficiency of the network and eliminate the tendency to throw away hardware sooner rather than later (I mentioned the climate already).
I can say confidently, that we are not at what I would consider the metaverse, it certainly does not need AR/VR devices and it is not run by the company calling itself “meta”. I know, that we have a lot ahead of us to make the next leap, but I am also confident, we will find these ways, evolve as a society and generate a lot of benefits along the way for everyone involved. I especially think, that we need to first build a world-wide digital society without borders or walls (however great or firery they are).
I also want to make it clear, that already any advance in the fields mentioned, is hailed as the arrival of the metaverse, but think about it: If you only advance in one of them, you are limiting the possibilities greatly and increase the risk for digital adversion and failure at the same time.
Say you advance AI but do not advance the understanding of data ownership and privacy. Or you commoditize scanning tools without make data storage and computation practically free. Each one innovation will be a solid step in its own right, yes, but the risk of breaking the digital economy is high as well, as monopolys or user adversion might be the result.
The metaverse as described here will not come as a sudden revolution and it will not be driven by a single company. It will be a gradual process a transformation of what we call www today, but it will nonetheless be vastly different from what we know now.