A specter is haunting the Internet -- the specter of decentralization. All the great powers of the old Internet have entered into a holy alliance to exorcise this specter: President and Prime Minister, Zuckerberg and Bezos, Apple and Samsung, clueless boomers and TikTok-obsessed zoomers.
While this introduction was stolen from the opening lines of the Communist Manifesto with a few relevant words and names replaced, it could just as easily have come verbatim from any of the approximately 4 million so-called "white papers" that have been published since 2009 describing some new cryptocurrency, blockchain, metaverse concept, or whatever else is certainly going to become the backbone of what people are calling Web3 (or Web 3.0). You see, we're on Web 2.0 right now, and Web 2.0 has largely failed to live up to its promise of a freer, fairer Internet. The problem with Web 2.0 is that it took the rugged individualism and Wild West mentality of Web 1.0 and threw it all in the shitter for the purpose of consolidating, collating, corporatizing, and commercializing what had previously been an amazing medium of human exchange. As of 2023, the Internet is essentially dominated by a handful of massive corporations like Facebook, Google, and Amazon. The core of this digital Gleichschaltung can be summed up in one ugly word: centralization.
If centralization is the disease, then what is the cure? The answer, obviously, is decentralization. Now you might be thinking to yourself "if centralization is what mainly separates Web 2.0 from Web 1.0, does that mean that Web3 will basically be like a return to Web 1.0 with better porn streaming?" No! Web 1.0 was filled with what were basically unchanging, uninteresting websites that could be built in the space of a day with minimal knowledge of HTML. The barriers to content creation were thus not particularly high, but the majority of the people who used Web 1.0 were passive viewers who basically treated the Internet like a library or a television where the screen had a lot of subtitles.
In the mid-2000s, Web 2.0 inaugurated significantly more complex website designs that raised the aforementioned barriers to the point that most people could not easily participate in website creation, which in turn led to people flocking to platforms where most of the work was already done for them, leading to a boom in social media use driven by services like MySpace, Facebook, and YouTube. You didn't really have to do anything other than fill in a few blanks on your page(s) and upload your content. Web3 understands that people fundamentally like the idea of being active participants in the digital space, but they don't like being beholden to a very small number of large corporations (and by extension the extremely small number of people who control those corporations). We can't go back, therefore in much the same way that True Communism addresses (or at some unknown point in the future will address) the ills of modernity with a rejection of the conditions that led to modernity, so too will Web3 address the ills of Web 2.0 with a rejection of the conditions of Web 1.0.
Back to Reality...
At this point, I need to abruptly change my tone because otherwise I'm going to start buying into my own ironic hype. Hype is actually the underlying factor behind everything that will potentially go into Web3. I say "potentially" because (a) Web3 does not currently exist and (b) there is no clear consensus on what exactly Web3 entails other than the nebulous idea of "decentralization." As you might have guessed, this theme of decentralization in the discourse surrounding Web3 is inextricably linked to "decentralized" digital fads like cryptocurrency and NFTs. I'm going to assume anyone reading this has at least a basic understanding of these concepts, mainly because I don't want this to become a book-length explanation of them. However, I'll still give some background and context as necessary.
An idea that is related to Web3 is what is called the metaverse. It's probably not really correct to refer to it as "the" metaverse since there can be a theoretically indefinite number of them, but I digress. A metaverse, on its most basic level, is a type of 3D graphic user interface for the Internet featuring user avatars and virtual representations of things like stores, offices, houses, etc. (There is some debate as to whether virtual reality and/or augmented reality functionality ought to be necessary for something to qualify as a metaverse.) Instead of driving to Wal-Mart in the real world, your avatar can walk to Wal-Mart's storefront in the metaverse and go up and down all the virtual aisles and do all your shopping from there. Presumably, the items that your avatar purchased from meta-Wal-Mart will show up at your actual house (and perhaps you'll get some digital items for your meta-house too).
There's a good chance that reading this has left you a little confused. You're probably thinking to yourself "but I can already do that by going to the Wal-Mart website, typing in what I want, and having it delivered to me; meta-Wal-Mart sounds exactly the same except I have to go through the extra step of playing a video game to get my paper towels and cheese dip." And you would be correct! But this meta-Wal-Mart is integrated into a world with games, social media apps, streaming services, and any other digital asset that you can visually represent in a 3D environment; imagine being in the middle of playing your favorite MMORPG and realizing you didn't have any food in the house, so your character can ride his dragon to the Pizza Hut kiosk, order something that will probably taste like shit, and then go back to gaming in one seamless motion! That sounds good, right?
One of the inherent problems in critiquing a technology (whether it is real or hypothetical) is that doing so invites snarky comparisons to people who said things like "why would I want to take a phone with me everywhere?" or "I don't need a computer at work" or "talking pictures won't be around in 5 years." However, I would argue that whereas cell phones actually made people's lives easier (to some extent), the metaverse really doesn't. It's a lot more convenient to type "paper towels" in a search bar on the Wal-Mart website than it is to visit a virtual meta-Wal-Mart, virtually walk down the aisles, virtually pick up a roll of digital paper towels, and then virtually check out. "Surely they would just let you type 'paper towels' in the meta-Wal-Mart search bar, right?" Well, possibly, but again, what's the point? You can already do that on their website. Add to this the fact that the metaverse is not a new category of device or even an improved version of a current device and the comparisons to things like cell phones fall flatter.
While it might seem as though I'm veering slightly off-topic by harping on this metaverse thing, it's actually a necessary piece of background information before we go into what is simultaneously the best and the worst representation of Web3 that currently exists. It takes the philosophy of a decentralized internet to its (il)logical extreme and incorporates cryptocurrency, blockchains, NFTs, and more into a metaverse. Behold: Decentraland.
Putting Theory into Practice
Decentraland can best be described as "digital anarcho-capitalism in Second Life." I guess at its core, Decentraland is a video game, or perhaps more accurately, a video game platform. While it lacks VR/AR compatibility, it satisfies most of the other requirements for a metaverse. It includes aspects of gaming, social media, traditional media (e.g. streamable music and video), and of course commerce. It runs on the Ethereum blockchain, uses its own cryptocurrency called MANA, and digital plots of land are purchased with Decentraland's NFTs. What does one do with MANA? Buy stuff! Chiefly the things you can buy are more NFTs that relate to in-game (for lack of a better word) content like clothes for your avatar. What can you do with parcels of property in Decentraland? Build a house for your avatar and your avatar's stuff (which you have purchased with MANA) and "live" there...oooooorrrrrrr you can put a digital storefront there so other people can buy stuff with MANA! Or you could consider it an investment opportunity and sit on it for years and then sell the NFT for the property for more MANA!
Believe it or not, several actual real-world companies have storefronts there. And they're ones you've probably heard of: Atari, Hershey's, Adidas, and Samsung, among others. It's worth noting, however, that the Hershey's area was apparently purchased by the company's branch in India, and it's probably the case that many of the companies themselves do not actually own the NFTs, but rather their advertising agencies do, and they include "metaverse presence" as part of their package. Presumably, when a company drops the agency, the parcel of land is wiped and reused for whoever signs up next. A problem arises when attempting to interact with these storefronts, though: you can't stay in Decentraland to actually do anything. Clicking a "Buy Now!" or "Contact Us" button generates a prompt to let you know that you'll need to go to the relevant page via your browser. And while MANA might be the currency of choice for products and services within Decentraland, it should go without saying that when you visit one of these links, you need to pay for your candy bars and shoes with utterly contemptible centralized fiat currency.
You've probably noticed that I've spent quite a bit of time talking about the economic aspects of Decentraland. This is because there is frankly almost no aspect of it that is not commercial in nature. The entire platform is based on the acquisition of MANA and NFTs. Now of course, no platform like this could exist or sustain itself without some kind of consistent revenue stream; I have no problem with Decentraland needing to make money. It's even kind of cool that the userbase can (theoretically) make money. But compare Decentraland to something like, say, Roblox.
Roblox is a gaming platform geared toward kids. Users can create "experiences" that other people can play. I have no idea how many experiences there are, but it has got to be at least in the hundreds of thousands. The experiences range from utterly mundane running simulators to complex platformers to survival horror games. While every experience is free to play, players can buy an in-game currency called Robux that can be used to buy avatar content (which users can also make and sell), VIP memberships, and a host of in-game items for almost every experience. The content creators will get a portion of the Robux which they can convert into real cash (although the exchange rate is something like 30% of what the Robux actually cost, but that's not the point). Roblox has millions of daily users and generates enormous sums of money. Decentraland's active userbase is a matter of controversy, with generous estimates placing it around 8,000 on the high end and less optimistic assessments putting the number around 40 (not a typo) on the low end. Roblox is what Disney World would be like if it cost nothing to get in and there were a ton of other people for you to have fun with, but you could buy souvenirs, better places in line, better food, faster roller coaster rides, and any other number of little upgrades. While this would ultimately become annoying and expensive, the fact of the matter is that you don't technically have to buy anything at all to enjoy it. Decentraland is more like plunking down a bunch of money to rent a store in a severely depopulated strip mall and the "fun" comes from visiting all the other stores in the strip mall and spending money for the privilege of being able to...well, spend more money (which is not usable outside of the strip mall).
This doesn't really even address the other issues with Decentraland. The graphics are, to put it kindly, behind the times. The "games" that exist inside Decentraland are rudimentary and buggy and are not really fun in any conventional sense of the word. Decentraland has a news service that is ostensibly oriented around the happenings of the metaverse, but because nothing of note ever happens there, the "reporters" usually resort to livestreaming themselves playing games. Famous musicians (e.g. Megadeth and Grimes) have held concerts in Decentraland, but these concerts usually consist of some pre-recorded messages from the artists interspersed with previously available studio or live recordings. In 2022, Decentraland hosted the first Metaverse Fashion Week, which featured prominent clothing brands parading low resolution avatars wearing poor representations of their latest designs to an audience of perhaps dozens of people. Coverage of this event was...polite. I would assume that most of the fashion journos assigned to watch it really didn't understand what they were looking at, since most of the articles about it discuss the "potential" for future events as the technology matures.
Now to be totally fair, Decentraland's shortcomings -- particularly the aesthetic ones -- are not necessarily refutations of the metaverse or Web3. It could be argued, after all, that the ideas are still perfectly valid and attainable, it's just that Decentraland in particular has not realized them. Still, the references I made to communism earlier weren't just jokes or rhetorical devices: so much of the discourse around Web3 is extremely reminiscent of arguments that "Real Communism" has never been tried, and therefore the deficiencies of the Soviet, Cuban, Chinese, or North Korean systems cannot be used as arguments against it. This is not to suggest that Web3 is a digital type of communism, because all of Web3's "features" are motivated by a basically libertarian, anarcho-capitalist ideology. However, evangelists of both communism and Web3 are utopian to a fault; their systems will eventually solve every problem and people who don't see that are either uninformed or have a vested interest in the old way, and are therefore enemies who need to be overcome.
As I mentioned earlier, it is waaaaaay beyond the scope of this writeup to address all the issues with cryptocurrency et al, so what I'd like to do instead is look at some of the ways these technologies are supposed to address the problems with Web 2.0 and discuss whether or not they actually do that. I'll be discussing Decentraland here and there again, but most of what I say in this context will be applicable to core Web3 concepts overall.
Crypto and Blockchain
NOTE: If you are already familiar with the ideas behind cryptocurrencies and blockchains, you can skip the next 2 paragraphs.
Without a doubt, the best known Web3 concept is cryptocurrency, even though it is firmly rooted in the Web 2.0 era. The most basic definition of a cryptocurrency is that it is a digital currency that is not controlled by any central authority (e.g. a government) and that its use is not dependent on banks or other institutions to authorize or verify transactions. The first and most famous cryptocurrency is Bitcoin. Bitcoin was created in the wake of the 2008 financial crisis that crippled economies all over the world. While many people walked away from the crisis thinking that financial institutions needed stricter government regulation, others came to the opposite conclusion: that central banking and government interference in the economy were just as culpable as the bad actors who created the problem. While the merits of the latter argument are debatable, there is no denying that there was a widespread loss of faith in The System as a result of the crisis. If we can't trust the banks or the government, who can we trust? Nobody. What we need, then, is to remove trust from the equation entirely.
Enter the blockchain. The blockchain is a permanent public record of every transaction ever made using the cryptocurrency in question. If I want to give John 1 Bitcoin, we're not going to rely on a possibly duplicitous bank to assure John that I actually have the 1 Bitcoin, we're going to rely on the blockchain to tell God and the whole world that I do. But for the sake of privacy, you're not going to see the name "John" appear anywhere on the blockchain; you're just going to see a digital wallet named something like 0ced53txx09ol154z receiving 1 Bitcoin from another similarly-named wallet. We'll also have other people confirm the details of the transaction publicly, and for their effort, they'll receive a teeny-tiny commission for helping to facilitate the transfer. Once the transaction is confirmed, it is added to the blockchain, and it can never be edited or removed after that fact; it will be there long after John and I have ceased to exist.
This is oversimplifying both concepts, and there are a few variations on them, but you get the idea. So do crypto and the blockchain really help us reach the final stage of a decentralized internet? The answer is somewhere between "no" and "yes, but that's not necessarily a good thing."
Night of the Long Forks
In 2016, Ethereum -- which maintains its own blockchain and uses a cryptocurrency called Ether -- proposed the DAO: the Decentralized Autonomous Organization. This is a little confusing, because the concept of a decentralized autonomous organization predates its use by Ethereum and there are many different DAOs out there; for the sake of clarity, we'll just call this one "the DAO" and any others I bring up will be called (service)'s DAO. The best way I can describe the DAO is as a voting-based semi-automated venture capital fund with no official leadership. So let's say someone came to the DAO and asked for $1 million to make hardware upgrades that would reduce Ethereum's utility expenditures; the stake-holders would vote, and if the "Yes" vote won, the DAO would automatically distribute the $1 million. All of the code behind the DAO was open source and it was posted to the Ethereum blockchain. Ethereum raised an incredible $150 million in Ether for the DAO. The problem was that a number of security flaws in the code of the DAO were uncovered and published about a month before it launched. These issues were not corrected when the DAO launched in June of 2016, and $50 million -- 1/3 of the total value that had been invested in the DAO -- was stolen. This was not just 1/3 of the value of the DAO, this was somewhere between 5% and 10% of all of the Ether in existence.
Ethereum had a choice: maintain the integrity of their blockchain and let the theft stand or reverse it and create a hard fork in the blockchain, which would have the effect of making two blockchains. One chain would recognize the theft as well as all the transactions that came after it and one would delete the theft as well as all of the other transactions. Neither option was particularly good; keeping the theft would diminish investor confidence but reversing it would probably upset the most devoted members of the userbase. Ultimately, the decision was made to reverse the theft and create the fork. Naturally, this created a schism in Ether, with Ethereum Classic retaining the theft and Ethereum removing it. Predictably, the value of the Ether that stayed with Ethereum Classic crashed.
While I lean more toward the opinion that the decision to reverse the theft was the correct one, it was 100% not the action of a "decentralized" blockchain. There's also the fact that if Peter loses $50,000 worth of Ether, he's just shit out of luck, but once it's $50 million of what is effectively a VC firm's money, they pull out all the stops and violate one of their most sacred tenets to get it back. So while it's no big deal for some random people to lose everything, there are apparently certain crypto institutions that are too big to fail. As much as I hate to paraphrase Karl Marx again, all great historical events occur twice; the first time as tragedy, the second time as farce.
Now to be fair, hard forks are generally thought of as a measure of last resort. In 2014, the NXT platform experienced a massive theft similar in scale to the Ethereum event, but the userbase decisively rejected the option of a hard fork. Instead, a ransom was paid to the guilty party and most of the Nxtcoins were retrieved. And while I suppose this is more ideologically consistent with the notion of "decentralization" than a hard fork, it rewards really shitty behavior and screams "we do, in fact, negotiate with terrorists." How would this tactic benefit anyone at all in Web3? How many serious real-world businesses -- which are vital to the success of the metaverse concept and Web3 overall -- would hang around after something like this?
It's a great irony, though, that the king of cryptocurrencies has gone through many hard forks for significantly dumber reasons. We could argue that Ethereum losing $50 million is a state of emergency that necessitates suspending the normal rules. Bitcoin's hard forks, by contrast, are more akin to the ideological disputes among the communists, socialists, anarchists, anarcho-syndicalists, and moderates that contributed to the republican defeat in the Spanish Civil War. In August of 2017, the question of transaction processing time led to a hard fork with the offshoot becoming Bitcoin Cash. Two months later, there was another hard fork in the Bitcoin blockchain over the processing requirements for mining Bitcoin, leading to the creation of Bitcoin Gold. Toward the end of 2018, an argument in the Bitcoin Cash community over maximum block sizes led to the fork getting a fork called Bitcoin SV. Bitcoin Cash would get yet another fork in 2020 called eCash over a similar issue.
All of these different iterations have their own versions of Bitcoin that are not compatible with one another and all of them are worth wildly divergent amounts of fiat currency. These Bitcoin hard forks are not just failures of decentralization, they are actively harmful to the purported goals of the cryptocurrency. Imagine if there were hard forks in the Euro and a €20 note printed in France was worth less and accepted in fewer places than a €20 note printed in Ireland. Regular Bitcoin already struggles to be accepted as a legitimate medium of exchange due to its own inherent volatility, how in the hell are you going to explain to a retailer "no no, I have Bitcoin SV, the true Bitcoin!"? Ironically, a big enough real-world business with a Web3 presence could conceivably take on a role similar to that of a central bank by refusing to accept any cryptocurrency that does not meet certain stability benchmarks, which would likely precipitate a mass exodus from the others and in turn cause other Web3 businesses to only accept the larger company's preferred coin(s).
As I mentioned earlier, the DAO is not the only DAO. A key component of most DAOs is the smart contract. Smart contracts are basically coded in such a way that they automatically execute once a certain event has occurred. See, the problem with telling a person to do something is that you're reliant on that person actually doing it. Smart contracts remove the human element. In 2021, for example, Decentraland's DAO took up the question of banning the name "Hitler" from the service. Oh yeah, I forgot to mention earlier that it's possible to create usernames as NFTs in Decentraland for the purpose of (what else?) buying and selling them. In this case, a majority of votes agreed that Hitler should be banned, but because not enough people participated in the vote, the smart contract was not executed. Or I guess it didn't execute itself. Whatever. We'll get back to smart contracts in a bit.
You'll notice that I've been using some awkward phrasing when it comes to talking about votes on things. The reason why is because votes in Decentraland and many other Web3-adjacent communities are not based on the principle of one-person-one-vote. Rather, voting in these arenas is more similar to voting by shareholders in corporations: the more shares you own, the more your vote counts. In the case of Decentraland, Voting Power (VP) derives from MANA, names, and land. For every MANA you possess, you get 1 VP; for every name you own, you get 100 VP; for every parcel of land you own, you get 2,000 VP. Because the active userbase and therefore the number of potential voters of Decentraland is so small, it really doesn't take many whales (that is to say very high spending users) to ensure whichever outcome they'd like.
Now because this relates to Decentraland, the issues at stake are not generally speaking that important. However, you can see some really lopsided results. For example, there was a request from a user for a $200,000 grant from the Decentraland DAO fund to improve a "district" in anticipation of the next Metaverse Fashion Week that was rejected by a margin of 90% (2.4 million VP) to 10% (288k VP). However, 79 users voted "Yes" and 72 users voted "No," and of the users that voted "No," just 5 of them accounted for over 2.1 million VP. In absolute terms, yes, I agree that it would be silly to give someone $200k USD to beautify a virtual district in a sparsely populated digital environment that struggles to perform at the visual level of the original version of World of Warcraft. However, the fact remains that this DAO VP concept has done nothing more innovative than recreate the electoral system of the ancient Roman Republic.
I don't think it is too difficult, however, to imagine a scenario in some other Web3 setting where this type of weighted DAO voting could produce profoundly negative outcomes. Any proposal to change the voting from VP to one-man-one-vote would naturally be defeated by extremely wide margins. But what if you had a small clique of high-VP users who, by virtue of their financial assets, decided one day that all incoming traffic from a particular country ought to be banned? What if this same clique decided that all users below a certain VP threshold lost their voting rights? There are alternatives to this type of weighted DAO voting, including the concept that you can spend a certain amount of banked VP on each vote; in other words, if it's an issue you don't feel too strongly about, maybe you'll just spend 1 VP on it, but if it's something that really matters to you, you'll throw all your VP into it. On the face of it, this seems like a decent alternative to a permanently entrenched ruling cabal, but let's be real, if you could afford to put $100k into something like this beforehand, you could probably do it again. The guy who maybe only has 20 VP will expend all of it and thus remove himself from the political process much, much sooner than someone with 10,000 times that amount. These weighted voting systems might not "centralize" power in the strictest definition, but they certainly consolidate it.
All of these decisions, of course, would be carried out by smart contracts. While there is a certain appeal to efficiency in the use of smart contracts, their actual purpose is to attempt to reduce or remove liability for potentially undesirable decisions that the platform and its associated DAO make. DAOs, in fact, have the same underlying principle. The argument is that "well, no actual person took this action, so nobody can be sued, charged, fined, or arrested." The legal merits of this attempt to avoid responsibility for potentially tortious or even criminal actions are dubious, but I've heard claims that copyright-infringing NFTs are not actionable because once they've been minted, they're on the blockchain, and you can't undo the blockchain! If a Mickey Mouse NFT is procedurally generated and minted as a result of actions taken by a DAO and then executed by a smart contract and then sold for any amount of money exceeding $0 USD, I guarantee Disney is going to pursue every person and piece of machinery associated with this action until the only place in the world they'll be safe from subpoenas and judgments is in a server room in a bunker 20 feet below the streets of Pyongyang. If you piss off the wrong person or company over copyright issues, you're going to start seeing enforceable court orders to hard fork blockchains, regardless of whether a DAO and a smart contract were "responsible" for the infringement.
But let's move this idea (further) into the realm of logical absurdity. You can go on YouTube right now and find videos of commercially-available hobby drones that people have modified to hold and fire handguns. As a Second Amendment enthusiast, I think this is absolutely one of the dumbest, most dangerous, and least responsible things someone who purports to be a supporter of gun rights in America can do, but it can be done. Let's say I build one of these, but I relinquish control of it to a DAO. Someone creates a proposal asking "should Sol's gun-drone be used to shoot and kill (insert person here)?" While 90 people in the DAO vote "No," they only have a combined total of 10,000 VP; 10 people with a combined VP of 4 million vote "Yes," and a smart contract executes a command to have the drone use GPS to fly to this person's house and shoot them. Was this a murder? Well, murder implies a murderer, and since this was an entirely automated process, there's no "murderer" to speak of. You could say that the person who made the initial proposal was the murderer, but what if their proposal was filled with sincere begging and pleading not to do it? What about me? I built the thing, sure, but I gave it away to this DAO years and years ago. What about the DAO overall? Sure, only a few people in the DAO wanted this to happen, but we don't only punish individual bad employees at corporations, we punish the whole corporation (in theory, anyway). So what about the people who voted "Yes"? What about the people who own the server that hosts all of this content? What about, what about, what about...?
You get the point. I know that the scenario I've just described is unlikely to ever occur. But again, the question is not so much whether something is likely, but rather whether it is possible, and whether in turn it is desirable. Most people, even DAO and smart contract enthusiasts, would (hopefully) say no to the final question. They could argue that there would be safeguards in place to prevent something like this from happening, and indeed that might be true, but as in the case of Ethereum's hard fork, something like that is not consistent with a philosophy of decentralization. Like every truly functional system in the world, there are exceptions to the rules, and (at the risk of sounding like Carl Schmitt) there needs to be somebody competent and empowered to recognize and resolve these exceptions. But it should be somebody whose judgment we can trust with a mechanism in place to remove them if they violate that trust too egregiously. It's almost like I'm describing a system of government with a central authority...
Holy shit is this waaaay longer than I intended it to be. It should be pretty obvious from reading this that what most people mean when they talk about Web3 is an even more monetized version of Web 2.0 with the difference being that the money goes to cryptocurrency fanatics rather than Facebook, Google, and Amazon. The problem Web3 seeks to solve is not re-democratizing the Internet through decentralization, but rather the historical injustice that all these other people and companies figured out how to become billionaires using the Internet before Web3 evangelists did. In fact, a common retort that these people use when someone challenges them, expresses skepticism about their ideas, or straight up FUDs (Fear, Uncertainty, Doubt) is some variation on "lol ok, I'll have fun making money, you have fun being poor." "Decentralization" in this context is what some communist theoretician somewhere will probably one day call "degenerated libertarianism."
Libertarianism is like communism -- and indeed virtually every other ideology -- not just in that it has a surface-level appeal that large numbers of people find reasonable but also in that it suffers from problems of scale and a susceptibility to manipulation (and the latter is mostly a result of the former). The ideas of "live and let live" and "don't have burdensome rules and restrictions" are attractive to most people. But what makes for a good ethos in a community of 20 people might not necessarily work for a city, a country, or an international computer network encompassing almost every facet of modern life. We can already see this in action if we say Decentraland is the city and Bitcoin or Ethereum is the country; I don't think it's worth trying the experiment at the third level.
I didn't even discuss a lot of the other things I wanted to get into, such as the fact that nobody really ever expects crypto to be a viable medium of exchange and most people instead buy it and promote it as a speculative "asset," or that smart contracts have been used to install and execute damaging malware, or the fact that there are somewhere between 9,000 and 25,000 cryptocurrencies (even the lower estimate would be more than all the fiat currencies that have ever existed in the history of the world combined), or how most NFTs are scams (I don't just mean this in the sense that they are overpriced and inherently worthless, I mean this in the sense that a large number of them were created solely as mechanisms for inflating the value of the cryptocurrencies associated with them and then quickly converting that now pumped-up crypto into filthy fiat currency).
If you really want to see what Web3 is going to be like and Decentraland won't run on your device, I invite you to check out Axie Infinity. Axie Infinity is kind of like a Pokemon game in that you have to collect these cute little creatures with different abilities to fight each other. You can even earn teeny tiny bits of crypto playing the game that you can convert that into fiat currency (but why would you do that?). All the creatures are purchased as NFTs, and the initial cost for a beginner NFTemon is about $200. You need a team of 3, so let's call it $600. If you can't afford that, no problem: you can apply for a "scholarship" where someone will sponsor you and you pay them back with the crypto you earn playing the game, and they'll even let you keep a little of it! This business model is called play-to-earn, and it is a perfect encapsulation of Web3. You have to pay money to get more money, and if you can't do that, a nice man will become both your boss and your landlord. Decentralization, folks!