Kevin Kelly’s Mirrorworld is missing one key piece: identity

What will life be like in the hyperconnected world that lies in our not-very-distant future? Kevin Kelly offers a compelling vision in the March 2019 issue of Wired. But there’s a key piece missing from his vision: digital identity.

In “AR Will Spark the Next Big Tech Platform. Call It Mirrorworld”, Kelly describes how converging technologies will create a kind of alternate universe that sits on top of our quotidian world, or perhaps alongside it, yielding a digital overlay in some cases, and in other cases, a complete digital replica of the real world. A mirror image that is rendered not in atoms but bits. Hence “Mirrorworld.”

It’s a good article. Much of it matches with my own experience in designing and launching dematerialized products, digital services, augmented reality, blockchain for supply chain, tokenized assets and digital identity. As a society we seem to be in the early stages of building a full-featured digital world that augments and extends the tangible one where we do most of our living.

Fans of “Ready Player One” may celebrate Kelly’s vision, as will those who thrilled at earlier sci-fi classics Neuromancer and The Matrix.

In my professional role as a strategist and advisor to companies large and small, I’ve been grinding away at the coalface of this particular topic for several years, and my book Vaporized covers some of the same terrain, particularly where I trace the power structure and value created in the digital domain. That’s why I read Kelly’s piece with a great deal of interest. Many of the things I’ve built, designed, proposed or written about are included in Kelly’s sweeping vision. I admit this might be an annoyingly solipsistic way to think about the topic, but somehow it also seems appropriate to find a reflection of one’s self in a place called Mirrorworld.

Kelly is always an upbeat and approachable oracle of technology’s near future. In Mirrorworld, he offers a view of the future a decade or so away that begins with simple enhancements like an AR overlay on a heads-up display (such as helpful arrows that appear on a windscreen to tell a driver where to turn) and annotations left like virtual Post-It on tourist destinations by previous visitors. Kelly calls these examples “trivial and elementary”, pointing out that unexpected potential will arise from the novel combination of AR features. The cumulative effects of those unanticipated combinations are much harder to predict.

The bigger vision in the piece, though, is what Kelly calls the “grand third platform” after the web and mobile Internet. According to Kelley, this next platform will consist of a complete digital replica of the real world. Kelly surmises that whoever dominates this next platform will be among the richest and most powerful people in history.

Correctly, I think, he points out that game designers will be the first to plumb the creative and expressive potential of this digital replica. As a game designer, I grok his drift. Gamers don’t need a user’s manual or guide book to figure out a new environment. They just get exploring and learn by trial and error.

Kelly also describes the efforts made by NASA, General Electric and Microsoft to develop “digital twins” for complex engineering systems. These are digital replicas that are so accurate they can provide a precise model of the actual wear-and-tear from atmospheric elements in a real jet engine, enabling predictive maintenance long before a part fails, and thereby an entirely new business model whereby GE can sell guaranteed uptime instead of selling a part. In my work with GS1 US, I’ve interviewed the folks at GE and Siemens and several other industrial giants who rely on such digital twins every day to maintain and operate jet engines, gas turbines and other big pricey complex industrial parts. In my own experience, the digital twin concept has gained way more traction among big industrial manufacturers than Kelly’s thumbnail sketch might suggest. At GS1 Connect, the annual gathering of supply chain innovation, we showcase digital twin initiatives every year and then we get into detailed discussions with their inventors.

The article is well worth reading, not only for the broad scope of Kelly’s vision but also his artful turn of phrase (“History will be a verb”). But I’ve got one bone to pick. There’s a missing piece.

Kelly doesn’t shy away from the potential issues and problems that might arise in Mirrorworld. He calls out several predictable issues, pointing out that “we’ll need mechanisms in the mirrorworld to prevent fakes, stop illicit deletions, spot rogue insertions, remove spam, and reject insecure parts.” And he expresses a vague hope for some non-governmental solution (i.e., no regulation) even as he underscores the likelihood that a new quasi-governing monopoly like Facebook or WeChat will arise as a byproduct of now-standard network effects.

But he stops a considerable distance shy of offering a solution. This is where the bright vision seems clouded by a blind spot. There’s nothing in Kelly’s Mirrorworld about digital identity.

Kelly nods in the direction of blockchain as a potential solution. But blockchain alone won’t solve all the problems inherent in a fully-realized digital replica. If we truly expect to have robot cars and other autonomous software systems that rely upon Mirrorworld to navigate the real one safely and reliably, then we will certainty that these virtual replicas are trustworthy, unhackable and secure. To achieve that, we’ll need more than the hopes and prayers of government officials or the pinkie-promises of benevolent tech dictators like Facebook.

To me, the solution will probably involve three core elements: identity, blockchain and a new kind of artificial intelligence.

Here’s how I imagine this will work. (Readers: Please help me out if I got this wrong!)

1. Blockchain. The key feature of blockchain in this context is that it re-introduces scarcity into the digital domain. It’s an axiom of our hyperconnected lives that digital objects can be replicated infinitely at no cost. That’s why the swift adoption of the MP3 file format instantly demolished the business model for the music industry: the ability to share my entire music collection with you while keeping it all for myself destroyed the value proposition in purchasing any album or single. But when a digital asset is recorded in the blockchain, its uniqueness is algorithmically secured: ownership by a single party is possible and title can be transferred to a new owner. These are the building blocks of value in a virtual item. Kelly correctly points out that blockchain is on the hunt for a job to do, and in Mirrorworld it fairly obviously will be used to establish and secure the scarce nature virtual assets like digital twins and digital real estate. But blockchain is not the whole solution.
2. Identity. To make this work, we’ll need a machine-readable identity system to ensure that each item is individually distinct and addressable by automated systems. Early blockchain-for-supply-chain trial deployments offer us a vivid illustration of the classic adage “garbage in, garbage out”. If your goal is to create a system that relies upon blockchain for tracking and recording the transfer of assets, that system will only be as reliable as the information that goes into it. To illustrate with an example: let’s suppose we’re building a blockchain system to track individual items in the global supply chain of organic produce. That system will only be useful if there is a trustworthy mechanism to ensure that the items it tracks are certified organic. That’s an identity problem, specifically related to provenance. Blockchain does nothing to address that part of the problem.
3. A more general sort of Artificial Intelligence. Ultimately this vast virtual world will also require a kind of artificial intelligence that does not yet exist. The most successful applications of AI to date are narrow. They are incredibly powerful in a very specific domain. Such AIs fail when they stray too far from their knitting. Mirrorworld will require a new form of AI that is not domain-specific and is able to recognize objects as they migrate across a variety of changing contexts…and it must also correlate these virtual items to their real-world counterparts. That’s a tall order. I have no doubt that such an AI will be forthcoming eventually, but I suspect it is a much bigger challenge than most folks realize. (I’d love to be pleasantly surprised to learn that I am wrong about this).

In my view, identity is the most vital piece of the puzzle. There’s no shortage of schemes proposed to manage digital identity, but to my knowledge there is only one identity system in existence that spans both the digital domain and the real world. That is the Digital Link standard proposed by GS1 as a new standard. Digital Link is intended to unify physical products with digital metadata. You might call it “Barcode 2.0” because it is intended as an upgrade to the original 50-year-old barcode standard.

The hundreds of thousands of manufacturers, distributors, retailers and marketplaces that rely on the durable GTIN system managed by GS1 don’t need to be masters of the digital domain. As long as they implement Digital Link, their products will be linked to all relevant digital metadata. This might include marketing material, video, instruction manuals, lists of ingredients or components, info about sourcing and provenance, details about shipping such as temperature or time in transit, product history, alerts about product recalls or upgrades. And it will also be linked to that individual item’s digital twin.

Crucially, Digital Link is machine readable. Which means it will support both human consumers and a fully automated supply chain in the future. And it fits right into Kelly’s vision of the machine-readable Mirrorworld. We’re closer to his vision that you might think.

My objection about identity is a minor criticism, more like a quibble. Kelly gets so much right, like the importance of making Mirrorworld machine-readable so that robots and automated systems can navigate in the real world by relying on an exact digital replica to detect the tangible physical objects around them, that it seems almost churlish to point out this one missing piece of the puzzle.

But it would be a colossal mistake to underestimate the significance of digital identity. To me, provably unique identity is the essential ingredient that will make Mirrorworld function properly. Without it, we will have duplicate worlds that blur and blend confusingly, a colliding namespace, overlapping and incompatible realities, virtual claim-jumping, digital squatters, identity thieves and even a new kind of blackmail whereby an unflattering or scandalous digital replica might be unleashed unless the rightful owner of an identity pays to reclaim it.

Your identity and reputation are the most precious things that you own in the real world. That will be true in Mirrorworld, too. And that’s why it matters that we spend time thinking (and debating) the best way to secure hybrid digital-physical identity for people, places and products. GS1 is far ahead in terms of framing the debate for product identity in the digital domain.

In my next article in this series, I will explore more deeply the importance of digital identity.

Previous piece: Let’s journal the rollout of IoT

Next article in this series: Complexity in Identity, Part One: the Bifurcated Personal Identity

+++++++++++++

If the topics in this article interest you, then why not join me for a discussion in person? I will be the host and master-of-ceremonies for the Innovation Track of GS1 Connect, the biggest gathering of supply chain experts in the world. If you are interested in blockchain for supply chain, business process automation, the application of artificial intelligence to manufacturing and retail, then this is a conversation you don’t want to miss. This year, GS1 Connect takes place in Denver Colorado on June 19 to 21. You can get early-bird pricing if you apply before April 15.

For 30 years, I’ve been focused on designing and launching new digital services. I supervised the launch of the world’s first mobile video services, some of the earliest PC games, online games and mobile games, and the biggest live online learning programs in the world. Today I serve as the Special Advisor for Digital Identity to GS1 US. GS1 is the global standards body for product identity. `