The author is confused. The point isn't to have every little thing decentralized (though there are plenty of tools in the space to do this if you desire). The point is so that different centralized entities can build on top on the decentralized data in a way that can be trusted.
If I want to build my own decentralized exchange allowing people to trade their tokens I can without the permission of anyone. If I want to create my own NFT marketplace to compete with Opensea I can without the permission of anyone.
Yes, it costs money to do things on chain. That's the part you sacrifice, the benefit you receive is the fact there is no single point of failure. If a crypto exchange goes down it doesn't make the price of cryptos go to 0. Even if they were all taken down simultaneously you could easily spin one back up pretty quickly. Web3 is simply a tradeoff of cost vs openness & robustness. Most things have no need for Web3, but it certainly has it's place for certain types of projects.
Web3 is full of hype, ponzis and fraud, but as a builder it's easy to get excited about what can be built with permissionless, programmable money and tokens.
"The only “utility” for a cryptocurrency (outside criminal transactions and financial frauds) is what someone else will pay for it and anything to pretend a possible real-word utility exists to help find new suckers."
This is simply incorrect. Just as a simple example: Being able to accept money for freelance work (whichever country you reside in) without paying Upwork 10-20% of your pay or relying on other intermediaries is "utility" no matter how you twist or turn it.
People who categorically dismiss cryptocurrencies or web3 as useful technology should probably check their biases.
This is not a good article. Or at least, very intellectually dishonest.
I could go point by point, but instead, just one will do. The comparison of ETH block vs Raspberry Pi. It is presented as if it’s some kind overlooked, unsolved issue by silly Ethereum developers. In reality, Ethereum is slower than Raspberry Pi by design, on purpose. It is meant to be runnable by a laptop and not taking the whole CPU.
The whole space becomes a lot more interesting once you begin to question, hmmm, WHY do these supposedly smart engineers make decisions I don’t get? There is almost always a WHY. And that is a lot more curious outlook than crypto-bro bashing.
And as a sidenote: why do we keep seeing these cheap shots at web3 and crypto regularly rising on HN? They rarely present any new insights.
They persist because there’s a lot of skeptics who only see the harms that crypto causes: making money laundering trivial with tumblers, making a safe haven for criminal activity by bypassing kYC laws, deriving much of its value from speculators who have no intention to actually use it as a currency, screwing over these customers with market manipulation that is illegal in most markets, making goods like graphics cards impossible to buy, putting a floor on the price of energy to be whatever can be mined, and contributing to global warming by spending the energy of a country on a silly number of transactions. But if you have a good article on why web3 is good For all parties involved, I will read that too.
80% of what the "skeptics" attack is based on issues that are not related to "web3" but to simple Proof-of-Work consensus mechanisms. PoW is acknowledged to be a sub-optimal system, everyone serious working in the space wants to get rid of it. Most of any new projects are already based on some other consensus system and at least Ethereum is poised to make the switch this year.
The other 20% of the issues are things that skeptics claim to be "causing harm", but either ignore to see that the harm already exists with the status quo (money laundering with crypto is actually really hard, and a lot more money gets laundered within the existing financial infrastructure) or ignore there are solutions to mitigate them being developed or in place.
There is no system that will be "good for all parties involved". Ever. It's all about trade-offs. It's fine if you are okay with the status quo, but don't go around thinking that everyone working in the space is either malicious or an opportunist.
> 80% of what the "skeptics" attack is based on issues that are not related to "web3" but to simple Proof-of-Work consensus mechanisms.
Not surprising when the vast majority of activity is on systems still using proof-of-work. Even the term "web3" is just a blatant attempt to profit off the success of the web as a technology. Not to mention, other systems that don't use proof-of-work have their own downsides. All of them are hugely inefficient compared to non-blockchain-based technology.
> The other 20% of the issues are things that skeptics claim to be "causing harm", but either ignore to see that the harm already exists with the status quo (money laundering with crypto is actually really hard
[Citation needed]
> , and a lot more money gets laundered within the existing financial infrastructure)
Again, what evidence do you have of this? And even it's true, it means nothing when the vast, vast majority of transactions happen on... existing financial infrastructure.
Are you really going to argue that a smaller proportion of crypto transactions are laundering money?
> There is no system that will be "good for all parties involved". Ever. It's all about trade-offs. It's fine if you are okay with the status quo, but don't go around thinking that everyone working in the space is either malicious or an opportunist.
Blockchain as a technology is a solution looking for a problem. It always has been. In the early days, you could be justified in thinking that there must be a worthwhile problem out there to solve with it, and that we just need to find it, because it achieved something genuinely new and exciting. To this day, not a single problem has been solved with blockchain. It just doesn't work as a foundation for a digital currency, and for so many reasons!
As for other uses of blockchain: ethereum is certainly interesting. Smart contracts are sufficiently powerful that it will be a while before I feel we've exhausted all possible uses for them, but at the end of the day, it's still in the "solution looking for a problem" phase.
In short: if you think you've found that killer problem where blockchain is the solution, go for it! I'm not judging anyone for working on it, and seeing where it goes. The issue is with everyone already invested trying to force these magic beans onto people who don't know any better, making outlandish and unsupported claims about a (so far entirely useless) technology. There's a word for this: scamming people. It happens to be illegal. If you own crypto, then you don't have an unbiased opinion on this: more enthusiasm for crypto is directly profitable to you.
- "The estimated amount of total money laundered annually around the world is 2-5% of the global GDP (USD 800 Billion – 2 trillion)"
- In 2009, total spend on illicit financial activities like money laundering was 3.6% of the global GDP, with USD 1.6 trillion laundered (according to the UNODC)[0]
> Are you really going to argue that a smaller proportion of crypto transactions are laundering money?
Yes! Without a doubt. We are talking about money laundering here, not "transactions volume that are not reported to any fiscal authority", which are two very, very different things.
> To this day, not a single problem has been solved with blockchain.
We can move value across the world in less than 10 seconds, without having to ask permission to anyone. That in itself might be no big deal for privileged folks living in places with (still) functioning institutions. But ask anyone who got blocked by Paypal or Google or even tried to make an international money order, and see if they don't think that crypto has solved a problem already.
> If you own crypto, then you don't have an unbiased opinion on this: more enthusiasm for crypto is directly profitable to you.
First, you can find a lot of things that I've written where I the biggest use case I promote is for stable tokens (i.e, crypto that is pegged to the USD or the EUR). The biggest use-case I am working on is to make payments easier, faster and independent of big centralized institutions [1].
Second, I don't "invest" in crypto as someone who buys lottery tickets and hopes to hit the jackpot. My focus is on collaborating to build an useful economy on it. You won't see me promoting/defending a project based on the token.
Third: Skin in the Game. Yes, I am working on the space so of course I want it to succeed. This doesn't mean that you have any right to say that I am being unethical or that I am not deserving of any of the potential rewards.
Your last three points will be solved on Ethereum by proof of stake. It's been in the works for a long time, but now the proof of stake network has been running in production for over a year, with $30 billion in ETH deposited. It's running in parallel to the main network, which is still mining, but the code to merge the networks is running on a public testnet. All their previous upgrades have gone live within a year of their public testnets.
In terms of "tps", transactions per second, Proof of Work Ethereum can handle 30 transactions per second.
Part of the development for "Proof of Stake" Ethereum includes scaling the number of transactions via sharding. Each shard will be able to handle 2,400 tps. As more shards are deployed up to 64, Ethereum will reach 160,000 tps. Please note, the upcoming improvement to Proof of Stake Ethereum does not yet use sharding, but it is under development.
As comparison 1,700 tps for the Visa network is a good benchmark. I recall Visa's technical capability is 10x or 100x that number though. This comparison isn't intended to claim "ETH beats Visa for currency stuff", but simply to give a sense of the potential impact of Ethereum as it improves. Thousands of transactions per second implies a powerful network, whatever it's used for.
These claims for Ethereum doing 100k tps on the base layer are bogus. It might look nice, but completely out of reality from a fully distributed network.
None of the core devs talk about these things anymore and everyone has accepted that the way to scale will involve layer-2 systems and some sacrifices in decentralization for performance.
Anyone claiming to do 1k+ tps on a blockchain is either lying or just running things on a very limited or expensive network. Even then, take a look at Solana and see how many outages they are having lately.
Hmm source? Unless something has changed since January 3rd, 2022, sharding is intended as I stated. Here's an interview with Vitalik one of the core devs of ETH: https://youtu.be/b1m_PTVxD-s?t=1466
Or perhaps you're saying Vitalik isn't relevant anymore? Perhaps he's only a figurehead now or something. Or perhaps I have something more to learn about how this all works. Again, sources would be great, I'm curious for what the future holds.
I am not questioning the roadmap, or that they want to have sharding. I am questioning the numbers.
Getting 2400tps per shard is already a pipe-dream, but tps does not scale linearly with the number of shards. The more shards you have, the more transactions will require cross-shard communication, bringing total tps down.
Also, the video you linked has Vitalik talking about roll-ups as the main way to scale. With roll-ups, sure, we can get to 100k tps, but like I said in this case we are not talking about base-layer anymore.
Ah ok I'm getting sloppy with the terms roll ups and shards and I fundamentally understand neither. Happy to read any articles you wish to share on this. You mentioned that the core devs have changed their conclusions, for example, would be great to hear what they're thinking.
Then, please-pretty-please do not put yourself in a position of spreading misinformation. Even if unintentionally. It's hard enough to fight the stigma even when armed with good information, when "critics" see unfounded claims like yours it confirms the bias against crypto even more.
> Happy to read any articles
No articles. Stay away from opinion pieces. Try to get actual documentation about the project. Ethhub has some good material about it (https://docs.ethhub.io/ethereum-roadmap/ethereum-2.0/eth-2.0...). Look at the "Layer-2 scaling" section to learn about roll-ups and the many other solutions.
It can do this because it doesn't rely on any sort of "work." It's just another internet protocol.
Proof of stake doesn't have anything to do with tps, so that won't change significantly. The solutions to that are rollups (second layers that inherit the full security of the main chain) and sharding (which multiplies the capacity of rollups). Some rollup systems are in production today, capable of a couple thousand tps. According to the most recent plan I saw, sharding would multiply that by twenty, initially.
Ethereum plans to scale is not just on the switch to PoS. It will also include layer-2 systems.
In any case, currently there are ~300000 validator nodes running the beacon chain, which does not require powerful hardware. For comparison, 13.4M units of the Playstation 5 were sold since 2020. 116.6M units of the PS4 were sold since 2013. I'd like to know the number of servers runnning Minecraft there are around, but you can unarguably say that people waste more energy playing videogames than with the ethereum network.
PS5 uses a peak of 200W[0]. If they were used constantly, it would be 200Wh/h or 1.75MWh/yr. 13.4M PS5s would kick that up to 23TWhr/yr. The energy use of Eth is estimated at over 100TWHr/yr[1]. So no… even if every PS5, a chip that has climate scientists worried, that was ever created was left doing something horribly inefficient (say, computing PoW coin) 24x7 it would be using a quarter of this one coin’s energy consumption.
Forget about PoW. I was talking about the 300k PoS validators on the Beacon chain ("ETH2.0"). You can run a validator on less than 10W. Any cheap celeron can do it.
It’s true that these limitations are known to the in-crowd, and also to everyone with some CS knowledge that bothered to understand a bit about the technical details.
But pro-crypto articles/posts/tweets usually fail to mention any of that, or hand-wave it away with an incorrect reference to Moore’s law. “World computer” is a specific phrase I’ve seen many times referencing Ethereum, and I personally find comparisons with the Raspberry Pi a very effective way to contextualize that.
As a consequence of all the selective information going around, some people are drawn into crypto that don’t really understand the limits of the technology’s capabilities. It’s important for the protection of those people to explain, in detail but in plain English, the gap between hype and reality.
Yes, sorry about that. I really wanted to find one conference talk that went into a lot of details about picking gas limit and why (sync catching up, etc), but after googling for a couple of minutes, couldn’t find it.
If a strong explanation is difficult to find by a proponent of crypto already familiar with the issues then Ethereum, Web3, and crypto in general have a serious communications issue.
No disagreement on this point: everything is so decentralized, with very limited “official” content, that providing “authoritative” resources is quite difficult. That talk wasn’t even in Devcon, if I remember correctly.
Average laptops are many times more powerful than a Raspberry Pi, and distributed computing with proof-of-work means nodes can contribute as much power as they have/want; no arbitrary design limits are needed.
Also the comparison was instructions-per-second between Ethereum's network and a raw CPU. Of course this trade-off in speed comes at the gain of decentralization and other features, however it remains to be seen whether that has really provided any net benefit overall.
Yeah, I think the truth lies somewhere in the middle. You're correct in that we have yet to see if there are any tangible benefits from a "social computer" as it were, and comparing Ethereum to any computer is like comparing a tree to an apple. They're fundamentally different things that exist to solve fundamentally different problems.
The "interesting" part of it comes from how we can benefit from small-scale, distributed computing in the long run. The fields history isn't very promising (NFTs had a lot of potential from a logistics standpoint; all we got were lousy thousand-dollar apes), but if people are able to align their incentives properly then I can still see a way forward for smart contracts. The issue is that there is an infinitesimally small chance that we will be able to agree on what it should be used for, and it's likely to languish in the same stupidity/scam hell that other crypto projects have fallen into. We'll either make a revolutionary step forward in trustless computing, or receive a rather expensive lesson about how cool technology does not a desirable product make.
Saying it's slow on purpose does not address the author's criticism unless you elaborate by explaining why that was the design decision and how the tradeoff is better than the author's alternative.
Sure, “to get rich”. Not a very productive discussion. The WHY I meant was “why these technical choices”. Even if we take a cynical view that all of crypto is purely a money extraction machine, we can still learn something new by looking at technical decision reasoning.
I feel like pretty often there's a new "crypto" toy, claiming that this time it's all different, this isn't a scam, and we need to stop talking about whatever was hot last week and grasp that this latest thing is definitely the future. We get plenty of those on HN too.
So you'd expect a counter-balancing flow of cynicism, and yeah, it's not going to have a lot of new insight because scams are more the same than they are different.
If you fell for a pyramid scheme where you're supposedly involved in property investment, and then years later fell for a different pyramid scheme where you're supposedly selling nutritional supplements based on organic fruits, you might feel like these are completely different business, why are the posts saying "This is a pyramid scheme" just reciting the same things each time when the property investment and the fruit pills were so different? Well because that's just window dressing, it didn't matter, the essence of the pyramid scheme was the incentive structure, the fact that, honestly, you couldn't make money from the actual "business" only from the pyramid. How the "business" supposedly worked was irrelevant.
As someone who generally agrees with the authors' "this system is silly" viewpoint, I completely agree about this article. It's very "preaching to the choir", while also having the sort of mean tone that would have undermined strong arguments, let alone the weak ones here. Quite disappointing from a well-credentialed voice.
I mean imagine this same person writing an article about Raspberry Pi: "you could buy a faster PC for only a few bucks more. These little computers are a waste of time." This kind of dismissal is perplexing. In pursuit of crushing web3 do we stifle curiosity itself?
So an engineer that made a system that was 1/5000 the speed and thousands of times the cost of a $45 computer wouldn't be a smart engineer. They'd be history's shittiest engineer ever.
Imo its clear that a lot of the critical discourse has radicalized to such a degree that it is essentially taken as a given that there are no smart engineers in crypto. You then obviously don't need to ask yourself this question.
Sam Bankman-Fried is the perfect argument against this. He went into crypto because of ethical reasons (to make a shit ton of money and donate it to the most effective charities and organizations, he's prominent in Effective Altruism circles).
I think he's the richest 29 year old ever? Maybe Zuckerberg beat him, but $22.5Bn isn't bad in 3.5 years!
I'm not so sure. When, as currently, that wealth is generated based on the losses of speculators, the conned, or simply poorly informed, then I'd class Bankman-Fried into the robber-baron class who sought to cement their legacies with philanthropic endeavors instead of their ruthless practices. The means by which they accumulate their wealth are the lens through which we should view their "good works".
I'm pretty sure - he was donating 50% of his salary at Jane Street prior to starting his current crypto endeavors and has pledged to donate basically everything over the next few years, of which he's already started. Initially staff at his companies had to donate regularly, 1% of fees from his companies are donated, and he advocates others in the space to do the same. The guy basically lives in his office eating fried veg and rice playing League of Legends in his downtime. That money is being made for charities and predominantly comes by enabling trading, which I don't believe is ethically wrong (his general good, in a utilitarian sense, is definitely higher than any qualms one might have against crypto anyway).
As of now, trading is mostly gambling with a side show of money laundering. Enabling it puts him on the level of a casino empire. I'm not inherently against gambling, but I also wouldn't consider a person to be a saint for plowing 1% of their casino empire's revenue-- or even most of its profits-- into philanthropy.
Which is sad because the best engineers I've ever worked with or met was due to this space. Still slightly stung by Diehl's article on 'The Haskell Elephant in the Room' that attacked a load of my colleagues prior to his edits last year.
I'm curious about this as well. Here's my simple mental model.
DNS lives on the core DNS servers and is replicated elsewhere. DNS being "down" would mean that the core DNS servers were taken down.
ENS [1] lives on the Ethereum blockchain. ENS being "down" would mean that Ethereum itself was down. ENS names (.eth) resolve to address hashes on the Ethereum blockchain.
So to answer your question, they are parallel systems, so one going down doesn't affect the other.
However, the thing I'm wondering is if DNS goes down, does it take ENS with it? I'm not sure how the whole ".eth" part works, is that on DNS.
Or even as simple as you don't have a working local resolver. But yeah, that's my basic question, is ethereum (and by extension ENS) dependent on DNS or could you run it all without any problems?
Why would anyone want to spend deflationary currency renting an ENS domain today when no one outside the echo chamber space cares about them and DNS is ~10x cheaper? The gas fees alone are brutal.
Oh I'm not making a case for why anyone would want to do any of this. I'm simply stating that the author is incorrect. You don't need centralized DNS on web3 because you can use on-chain ENS. And you don't need centralized storage or servers on web3 because you can use IPFS and Filecoin.
Yep, it's as expensive as all hell, I can't figure out why people will really want to use it that much either. But the author is still incorrect.
Front page of HN yesterday: Facebook/Google/PayPal/Apple/Whatever took down my page/email/account/app and irreparably harmed my business.
Front page of HN today: Why Web3 is a scam and has zero real world use cases beside grifting.
Surely this is getting old? Can we get an article about Ethereum straying from it’s mission of accessibility? About “low-fee” blockchains consistently grinding under load?
If blockchains are low bandwidth by nature how can the DeFi movement claim to be the future of finance? Do large VC allocations of projects purporting to be infrastructure threaten the future legitimacy of Web3? I don’t know, but it seems the only people that care about these questions are the frauds and suckers referred to in the article.
These keep popping all the time on HN recently and I don't like it. We had our fun taking jabs at people buying virtual receipts but nowadays it feels everyone wants to join the smug train. I'd appreciate some moderation on these articles (as in having much less of them around here).
Yeah, we know: people basically give money for a small receipt that says "this thing is mine", ignoring the fact that the paper of that receipt might spontaneously disappear one day. We got it.
So what? People literally bought paper strips absolving them from their sins some mere centuries ago. People do all sorts of stuff we don't immediately relate to.
IMO the pertinent question around the NFT (and the general Web3 / dapps etc.) frenzy is just one:
"If people want to buy stuff you don't like, well, let them, how does it harm you?"
If that's not properly answered and if you can't prove to me that's going to hit me in my wallet later on (and I mean no later than the next 2-3 years) then we're just taking turns of making fun of people whose needs we don't understand.
Example: I am not bashing anyone who is buying a Bentley. To me that's unnecessary and expensive but to them it's likely super important and that's the salient point here: you don't know these other people's needs and you don't understand WHAT exactly they are buying with their money. I would think they are buying an ugly overpriced car but they might be buying it e.g. so they are allowed in new circles which need demonstration of status?
I am posting this comment as a guy who single-handedly coded the entire backend of a Web3 app some mere 3 months ago and has seen the absurdity of people flocking to buy NFTs due to a good slogan on a website (hosted on Web 1.0 / 2.0 no less). But in the end I get it; a lot of NFTs are presented in the form of campaigns to support or champion a cause, and others are marketed at being unique and deserving of status, etc.
Good marketing, appealing to certain human traits, and people open up their wallets. Why not?
Not for me personally but I am not judging. In the end I am the guy who bought Airpods Max ($540) just so he can use them 2-3 times a week to listen to the occasional song or a podcast, so who am I to judge?
We all do things that are weird to others. Live and let live.
As for technical merit, the article is correct but IMO that discussion has been beaten to death already.
We live in the world proof of waste is polluting and most of us know people who are getting the sales pitch to buy in with money they can’t afford to lose. Those of us who work in tech are seeing the market distorted by companies which have no plausible path to viability (you’re still competing with them for hires and investors) and consultants pitching your senior management on huge gambles on unsuitable technology.
Fair points but that's hardly the buyers fault, no? The Bitcoin and Ethereum networks have been humming along for a long long time before the NFT wave hit.
RE: competing for hires and investors, hmmm. I do agree but I question how long will they last so it's IMO a self-correcting problem. I might be wrong though; I've witnessed companies with literally no product to sell coast on investor goodwill and general hype (to my amazement and dismay) for years.
I think it's a bit hard to separate the two, however: cryptocurrency had been developing a bad reputation and NFTs feel like they were ramped up as a marketing push to finally give people a reason to buy in. The large holders who want to recoup the money they put in have a lot of influence and media presence so any time you see someone like @cdixon talking up the benefits of web3 you have to remember that his company has a couple of billion dollars on the line and adjust your credulity to compensate.
I do think it'll eventually correct but never underestimate how long the market can remain irrational, or the damage which can be caused if they start getting connected in to other industries. There's a good argument that, for example, the real estate bubble was in part pumped up by the big sovereign wealth & pension funds who lost money on poor dotcom choices and were seeking new high-return investments. I really don't want to see that happen again if someone convinces them that they should go long on NFTs.
Well said. Is there some area of the internet / web / tech that's full of curiosity and exploration today? My answer 10 years ago was MAKE magazine, but that era is over now. Maybe people are just smug about everything these days.
There’s a lot of that going on still - the maker / 3d printing folks are still beavering away, there’s plenty of open source creativity, etc. There are three big changes I think make it harder to see:
1. Google nuked Reader at the same time Facebook declared war on the open web. I follow this kind of stuff on blogs but a lot of people definitely shifted to social media.
2. The field is maturing in some ways, and a lot of open source energy moved into companies. That means that small projects are competing with Facebook for contributors and attention, and that there is less low-hanging fruit. A cool wearable hack now has to do something an Apple Watch, etc. doesn’t.
3. A lot of younger people are worried about things like climate change. Clean tech seems to have a lot of the people who a generation ago might have been doing web things.
The problem is that all the blockchain stuff takes HUGE amounts of electricity and this is not without cost to society. The better analogy than indulgences would be rhino horn.
Yep, true, but again, can we expect the common citizens to grab their guns (which most citizens of the world don't have btw; USA is the outlier in this regard) and go hunt and shut down the cryptocurrency mining operations?
Clifford Stoll should write a new book dismissing cryptocurrencies and NFTs, etc. You know, like he did for home computers and the internet in Silicon Snake Oil (1995).
Stoll wrote that book in 1995 after the internet was noticeably changing the world. People were flocking to sign up with ISPs; businesses were seeing the benefits for product information, sales, support; people were using the web to research and book travel plans; email and chat were becoming quite popular (people had been meeting and getting married online since the BBS era); etc. It’s also important to remember that computers cost a lot more, network connectivity was even pricier, the software was primitive, and all three were incredibly slow by modern standards.
In contrast, we’re over a decade into the cryptocurrency push and the primary marketing hook is “buy now so you can sell it to someone else” with nothing anyone would miss if it disappeared. Unlike the internet, it’s been globally available to most of the world so you’d expect uptake to be faster rather than slower.
Your claim that the internet was noticable changing the world in 1995 is not accurate at all.
For example, if you look at the history of Elon's Zip2, the success of an online yellow pages wasn't obvious. Quote itself:
"“At that point, very few people were in the internet [business], so it was really a question of, ‘Is the internet going to succeed?’ Which we were huge believers in, and these guys were not.” "
> Your claim that the internet was noticable changing the world in 1995 is not accurate at all.
Look, I get that you need to find buyers for your tokens but lying to people who have first-hand experience is not the way to do that. I gave you a list of things real people were doing then, and it wasn’t even complete. For example, 1995 is the year RealPlayer came out of beta and you had adoption by major radio stations and NPR. Major news organizations like the BBC, Bloomberg, El Pais, etc. were starting to publish online. Amazon had launched the year before. Music fandom and downloads were picking up, lead by college students around the world tired of the mainstream selection by the record labels.
The fact that the yellow pages guys weren’t keen on their traditional business model being disrupted isn’t surprising. That wasn’t uncommon back then, but that doesn’t mean that the people who built their replacements weren’t getting started and finding plenty of people interested in using new alternatives.
The key difference is that cryptocurrency has taken a decade longer and has had almost no outside impact. Most of the things people are building are ways to spend or speculate on cryptocurrency, not doing anything you can’t already do or doing it better. If Bitcoin suddenly disappeared, no non-cryptocurrency business would be disrupted. Nobody who isn’t a speculator would notice a change in their daily life. Even porn sites barely use it, which you’d think would be popular given the US credit card processors’ crusade against naughty content.
Here’s a list of websites started BEFORE 1995 - note that they’re not just people talking about building or selling websites but rather normal things people like as part of their lives which were made better by being on the web:
https://en.wikipedia.org/wiki/List_of_websites_founded_befor...
No, it couldn’t have. I’ve given you a ton of examples of people using it for things other than selling internet access or software: all of those people went out of their way to get access because it let them do things they wanted to do, often buying expensive new computers to do so.
The cryptocurrency world didn’t have anything close to those barriers to adoption but still hasn’t come up with a pitch for doing anything most people want to do. There’s no equivalent of the hobbyists sharing photos and tips, businesses communicating better with their customers, people reading the news from other countries, etc.
> > businesses communicating better with their customers
> I'm sorry, but you are living in fantasyland. Those things didn't happen in 1995. Businesses asking for your email address in 1995... yeah right.
Again, why are you telling that things I personally experienced did not happen? Two of the three Southern California suburban high schools I attended during that period had email addresses and web access — one of the first public websites I built was for my high school library, linking to sites like Yahoo which would help you find what you were looking for, because it was already that useful for school. The corner taco shop probably didn't have email but an awful lot of professional businesses were starting to use it — a family friend's law firm was using internet email (following a decade of using CompuServe), that graphic designer, etc. 1995 meant Yahoo, Amazon, eBay had launched and were doing real business. You may recall hearing about Al Gore's push to connect universities with high-performance computing centers, too.
Here's that link again — go through the list and think about how many people were using these sites for things which were not related to selling internet access or software:
Technical users dominated, of course, but many of them were using the internet incidentally — downloading GNU software, for example, like the first Linux 0.9 kernel I ran — because it was better than what they were doing before. My grandfather is an electrical engineer and technical customers went online quickly because who wants to wait for postal mail to deliver letters and floppy disks when you don't have to?
Note how many are even further away from IT: NIH or astronomy projects, underground music, TV fan arts, members of the U.S. Congress, etc.
It's not just personal anecdotes — you have conspicuously ignored all of the links you've been given to major business and mass media showing adoption. I've tried to assume good faith but to be honest this is reminding me a lot of when a creationist tries the “god of the gaps” technique trying to ignore the overwhelming evidence that their claim was wrong and instead tries to pick away at whatever small detail they think is easiest to attack.
> I'm sure plenty of people have awesome anecdotes using crypto.
Oddly, you have been unable to provide those examples. Odd that, especially since your “big companies” list includes some like Microsoft who backed away due to volatility and others who immediately convert it into hard currency as part of the payment process. There is nothing world-changing about something like VISA except with higher transaction fees, and none of those businesses would care if Bitcoin disappeared because it's just one of the many options their customers have available for payment. They'd remove a button on their website which a tiny fraction of people use and that'd be it.
> In Dubai houses are getting bought using crypto.
Same deal: they're treating those as cash payments because the cryptocurrency is liquidated before the deal closes, because it's not legally approved in the UAE.
The Today show was the top rated news program in its timeslot in this era with 6.3 million viewers [1]. They even display and talk about their e-mail address (at ge.com!) even though they clearly had no idea what it really meant (not uncommon for morning TV).
Yes, actually. That year I was helping a family friend with his graphics design business build a website for it. He had a scanner for use in his print work - remember, Mac users had been using Photoshop since 1988 and it wasn’t the first to market - which we used to scan photos in addition to other work samples. It wasn’t cheap but it was routine - and compared to other pro photo equipment it’s not like cameras and lenses were free either.
You’ll also note that I said tips - even if photos took time to process and view, people loved sharing advice and work stories.
but... what is this stuff used for? the OP just gave you a ton of use cases for the Internet in 1995 which you ignored?
> businesses were seeing the benefits for product information, sales, support; people were using the web to research and book travel plans; email and chat were becoming quite popular (people had been meeting and getting married online since the BBS era); etc
For "crypto", it's... flipping coins to whoever pays more, doing illegal things and... apparently moving money to Argentina. Anything else?
Sorry to disappoint, but in 1995, almost nobody had any real benefit from email (except maybe some universities). If you think email was mainstream back then, your memory is playing tricks on you.
Look at the WHOIS registration dates for the internet domains of the Fortune 500 in 1996. Almost all businesses knew within a few years that it was valuable to them. Go on the Wayback Machine (starts in '96) and check out the various websites for General Electric, Toyota, GM, you name it. They were there. Here's a taste from the Fortune 500[1] in 1996:
So there you have it - 9/10 of the most valuable companies in the world had a domain name and website by 1995 (the only one missing being a tobacco giant). I don't think I would say "most businesses didn't believe it was relevant to them".
I don't know why we've moved the goalposts to 1994. Anyway, ENS has been available for five years (ethereum for seven, blockchain for 13). What businesses that are doing something other than sell their crypto project have an ENS domain?
You're accurate in that the plumber down the street, local dentist, bakery, etc likely didn't see the relevance of the web in 1994-1996. 13 years after bitcoin later I've been to maybe five of these types of businesses that take crypto or have anything to do with blockchain. 13 years after the release of the web there were over 85 million websites with a billion eyeballs.
Bitcoin didn't even show up on Google Trends until 2011. Two years after the web 9/10 of the top 10 Fortune 500 had a domain.
I'm doing my best to be respectful but how is this even a debate at this point?
1994 was just the only stat I could easily find, only trying to show the web was pretty small back then.
I don't think crypto is doing anything useful. It's been around long enough to have had important things (other than trading crypto tokens) built on it.
I just don't think most businesses saw the relevance of the internet in 1995, but it certainly grew fast and kept growing.
Not trying to be disrespectful, I just think we may be talking past each other a bit.
I think I generally agree: “most businesses” would include the guy who mows lawns or repairs shoes, but I think I would draw a divide for professional or especially technical services — they might not have been heavily online but I think they were starting to think that was becoming expected and if they did anything exchanging non-massive files there was a big time advantage to not paying FedEx or a courier to take floppies or CDs around town.
Heh. 1994 was when I helped convince the large technical service business I worked for that we should have an internet connection.
The big use case was downloading drivers and other things that previously had to posted on floppy disks, which often took weeks.
We also got web browsing, but there really wasn't much up there to begin with. You could play guess the domain name or find a few link listings on some pages. Altavista appeared pretty soon though.
Yeah, we take downloads for granted these days but FTP sure freed up a lot of waste time compared to filling floppies (remember people juggling how to pack them since each one cost enough money to add up?) and delivering them, or, later, having burn a new CD with updates for a handful of files.
Not only the guy who mows lawns or repairs shoes. Most small to medium non-technical organisations didn't use it. I did technical support for all kinds: hotels, publishing houses, printers, warehouses, charities... They were often installing leased ISDN lines to link their branches.
Definitely — I was mostly thinking of what the breakdown was for number of businesses versus the number of employees or sales volume. Starting around 1998, we had a client (spa.com) who was trying to build an online info & booking site for spas and you had a huge divide there with some spas being relatively large, stable operations where they had a T3 to a large facility with a ton of rooms and others being a 2 room operation out a dirt road in New Mexico with a maximum-length phone line barely capable of receiving faxes. On paper they both count as a business.
Anyone comparing cryptocurrencies, blockchain as an overall platform, etc to the internet should acknowledge a basic fact:
At 13 years from release the web had a billion users.
13 years from release all blockchain networks combined have 100 million users at best.
In an era with ubiquitous internet access, social media, mobile devices, etc new internet based platform adoption rates should always accelerate. For example in 13 years Facebook had 2 billion users.
In the modern era blockchains have 1/20th the adoption rate of a social media site should tell you everything you need to know about the reality of this space - the average person doesn’t see any value in it other than speculative trading.
The internet is OSI model [1] layers 1-4. Layer 3 is literally called the Internet Protocol. Layer 4 is TCP, UDP, etc.
Web is five (some would say six) and up and was released to the public in 1993. This is well documented.
Blockchain is five (some would say six) and up and was released to the public in 2009. This is well documented.
The OSI network model gives us a way to disambiguate "the internet" by it's component parts and dependencies. The web depended on the physical and network infrastructure that is the internet, as did blockchain.
The web was accessible to at best 15 million people worldwide that had internet access when it was released in 1993. Within 13 years 15 million internet users became 1 billion internet+web users. As others have pointed out for the majority of this time period there were many challenges to adoption yet the web was so clearly and obviously compelling people overcame those challenges to access the tremendous value and utility provided by the web.
Blockchains have been accessible to billions of existing internet, web, social media, and mobile users since release and yet maybe 10% of these users are using blockchain for anything at all.
I try to assume good faith but at this point the fact you're still equating the internet and the web, not acknowledging the drastic advantages blockchain had in 2009 vs 1989/1993, etc is extremely suspect. The web was created at CERN and they have an exact date for what they consider to be the birth of the web: April 30, 1993 [1]. When it comes to the web I'm going to continue to use their date.
For kicks let's go with your date of 1989. 13 years later in 2002 the internet/web had 587 million users [2]. Anything blockchain related has 100 million.
I can't understand how even without your understanding the difference between the web and the internet, my being generous and using your (wrong) date, and yet the internet still growing at well over 5x the rate of blockchain helps your argument at all. I don't believe your ability to understand this is the result of stupidity or ignorance - I think it's the kind of blindness and rationalization that can be brought on by faith and greed/financial self interest.
In all seriousness and with the utmost respect - how much crypto do you hold?
I hold crypto because I believe it has potential. That means it has a chance it will lose out against what already exist, but it also has a chance it will replace a lot of existing things.
What I hate about people here on HN is that they are sure crypto is a fad and will go nowhere. But at this point, nobody can tell. But of course, plenty of people here seem so sure about themselves.
The web left CERN in 1991, so it can't be earlier than that.
The internet was earlier, but there are two other things to remember: in 1971, a computer cost as much as a house, filled a large room, and required a professional operator. The first microprocessors actually came on the market that year and it took time for people to build devices around them or for those devices to hit the mass market. When that happened, you were getting things like text interfaces and keyboards, tiny amounts of storage on cassettes or floppies, etc.
Similarly, in the United States the phone company monopoly and expensive but slow modems meant you were looking at 300bps connectivity tying up an expensive phone line and communicating with someone outside of your immediate area was expensive per-minute charges.
That got better but it really wasn't until the early 90s that GUIs were mainstream and it wasn't until the mid-90s that computer costs had come down enough to make them affordable to the average person. Similarly, modems were pretty slow and limited so the early web was held back by the fact that a lot of people would need 5 minutes to load a web page with a fullscreen 800x600 image.
Demand for the web is what drove a lot of non-techies to buy a computer capable of running something like Windows 95, because they'd heard of things they could do on the web — whether that was educational research, getting software updates without waiting for floppy disks in the mail, following sports, connecting with other people who shared the same hobby, making travel plans, downloading porn, etc. the first web browsers and email clients were driving a lot of people to buy computers so they could use them. ISPs had sprung up all over the country and faster options like DSL and cable modems were game changers for the experience.
This is a start contrast to blockchain technologies which were close to globally available on day one — even in poor countries, there was at least someone with a network-connected PC — but despite a much longer period of time still haven't found much demand because they don't improve the average person's life unless they enjoy speculating on their ability to resell tokens.
Thanks, that's good to toss in things like Diner's Club and Facebook, too.
I find it interesting that you're bullish about the space but seem to be one of the few people asking the most important question: why should anyone join? I'm a couple of years older than you and it's been really weird feeling like that's strangely absent — things like CompuServ, the web, social media, smartphones, etc. all had such clear answers for that that people would be asking for help signin up because they knew they wanted it because it'd help them do something (school, trading stocks, gaming, reaching customers, etc.) but weren't sure the best option or something like that. It's been really weird to me that the answer remains nonobvious. I'd be happy even if it was as simple as “we want to turn PayPal into a smoldering hole in the ground” because a) such a deserving company and b) anyone who hears it would know immediately whether it's something they'd like — and that means that people in the space would know they need to hit certain cost and ease of use thresholds.
Do you have any references for other people, articles, etc with similar questions? They seem incredibly obvious to me but it's very strange to feel so alone in what I consider to be such an obvious perspective.
That's just it - people went through A LOT to get on the web in the early days but the value was so immediately compelling and obvious they considered it worth the effort. As I note in the rant corporations were snatching up their .com within a year or two. How many legit companies have their domain and presence on ENS (which itself has been around for almost five years now)?
A comparison of ENS and .com seems super interesting. In another comment you mention the difficulty of comparing chains vs popularity etc. Or comparing market caps and sich is so weird, what's the "market cap" of USD?
Whereas for ENS you have something real to compare to with .com, including registration data. ENS does seem to be getting some adoption mostly at an individual/celebrity level. Here's the leaderboard of top ENS users by # of Twitter followers: https://ethleaderboard.xyz/
I also think the ENS community in general is curious about the questions you're asking. Fundamentally ENS/DNS is pretty boring, so it's hard to get too wildly speculative.
You’ll note I specially avoid any of the financial aspects of blockchain (market cap, etc). I really don’t want to open that can of worms.
Chains are difficult to compare popularity because of the anonymous nature and how “users” are considered. How do you translate transactions per day and active addresses into users? Almost all users have more than one address. Transfers from hot to cold storage for a single user count as a transaction and involve multiple addresses, as one example.
Bitcoin is 13 years old. Ethereum is seven years old. ENS is five years old. Yet I can’t go to Toyota.eth and learn about their new models or whatever like I could with toyota.com in 1995 (18 months after the web was released).
Note I use the Toyota.com example because when my sister and I first got the internet in rural Wisconsin in 1995/1996 we used to watch TV with a pencil and paper to look for the fine print at the bottom of the screen in hopes for a new web address to visit. Toyota.com was one I remember writing down (for some reason). We thought it was so neat you could see something on TV, type a few letters in the computer, and get “taken” right to it immediately.
Pick a handful of $bigcorps you can think of and hit the Internet Archive Wayback Machine. You’ll find out the vast majority of the Fortune 500 had a website up at a .com within a few years (archive starts in ‘96, I believe). Not to mention the millions of users that had been building websites on geocities and elsewhere since ‘95.
By 2006 the web had over 85 million websites. In 1998 (using five years of ENS) there were over 2 million (including Google). There are currently 247 thousand ENS owners. 13 year old platform. Five year old service. Once again a fraction of adoption when compared to the web over the same time span.
Don’t take this the wrong way but a leaderboard of celebrities with Twitter accounts that starts with Paris Hilton isn’t going to change any minds on the credibility of the platform. Rather the opposite. Frankly anyone who thinks Paris Hilton being the leading proponent of a five year old technology platform is a good thing is either desperate or delusional.
It validates the huckster hype perspective. She was on late night TV recently in a very awkward “interview” promoting NFTs. Many have commented it resembled a video one might see released by someone in a hostage situation reading a propaganda statement [1][2][3].
In any case I’d be more interested to see an analysis of those top accounts and what percentage are actively selling NFTs or any other crypto-something. I’m betting it’s extremely high. Otherwise, there's always the Fyre Fest angle where most celebrities will sell anything with no due diligence or scruples whatsoever as long as the check (in fiat, of course) clears.
ENS is not boring - having an ENS address is a marketing tool to the cryptosphere for whatever you’re trying to sell them. This is one area where it’s identical to DNS.
Yes agree re: avoiding pricing and focusing on utility like ENS.
Your examples about web dev in the 90s got me wondering "what does Coinbase do today with ENS?". They're fans of ENS, surely whatever's great about ENS, Coinbase must be using it.
Well, as of today their ENS profile isn't anything radical [1], just an address, image, and link to their ENS delegate page.
That gets me thinking more about toyota.com. Do you think they had internal sites and computer resources before their public website? Just makes me wonder where their "toyota.com" site came from, was it as simple as "they heard about the web and then BAM immediately built a website". Wary of doing mental gymnastics myself tho, maybe this all just does mean crypto stuff is overhyped and not very useful.
Yeah it's a weird vibe. Imagine if in the early CompuServ days the general person knew about it and thought it was a scam only for scammers, terrorists, and creeps.
As a user, I find it fun to chat with other crypto peeps about what they're doing...but talking to the outside world seems challenging. I can't really explain how "CompuServechain" isn't just a waste of time...but I'm having fun so I just keep doing that. I got a site setup on IPFS and connected to a .eth domain!
Maybe the difference is now people aren't coming to you for help with crypto, whereas maybe they were with CompuServe and smartphones? So you're playing a different role.
The only requirement I know is that it's directed through a .eth, so the site itself may be boring. But most sites in the index seem to be hosted on IPFS and some even ask for a wallet to have app functions.
There were certainly people that thought the web was only for weirdos and nerds. I got bullied in school because I was on the web and playing with computers instead of playing sports, etc. Nerds weren't always cool. Life as a nerd got much easier when the money started flowing and it was completely obvious to EVERYONE we were taking over the world.
Thanks for providing the link; the popular selection[1] is most interesting. The entire space has gone so meta (a little on the nose) most don't even realize it... Imagine if all the internet offered for the first 13 years of it's life was different ways to sell you more internet. That's what the entire space looks like at this point and this link is yet another example - there are maybe three instances of destinations that I could see being even remotely useful, interesting, or usable to anyone outside of the cryptosphere.
I've been really excited about the space since at least 2017. After five years without a single "I can't live without this" moment that excitement has waned significantly.
For me that moment is ENS and IPFS which I just learned about a few months ago. I think it's so cool to have a decentralized website and URL and I never want to go back. Had used Tor addresses previously for the DNS part.
I realize this may seem like moving the goalposts -- but have you thought about comparing to the development of the internet instead of web? Arpanet 1971 means 13 years in is 1984. I figure you've thought this through so will have some good points for why web/1991 is a better analogy than arpanet/1971.
All the stuff you wrote here is really good. Could be a blog post instead of these more ephemeral comments.
Agree with your assessment that 100 million is an upper bound. I am curious about a deeper analysis as well that involves defining what "real users" are and then seeing how many transactions/wallets meet that criteria. i.e. of that "100 million", how many of them are simply buying and hodling? How many have done a non-exchange transaction (i.e. bought a thing, not coins. Buying NFTs counts! Or a pizza, or buying software, or staking or something)
The comparison to Diner's club is also relevant. 20 years from zero to 51% of all Americans.
Just getting to your "Why has blockchain user adoption been so terrible and what can we do about it?" section and will edit after that.
Edit: Well worth the read. If anything, the title and top image were a bit clickbaity, but the article itself was great, all the way through to the end.
Overall, the key question you point to is "Why did the web grow 10x faster than blockchain? Facebook 15x, and mobile 35x faster?"
I'm curious about this! Is there a recent technology that did grow at 10x over 13 years that we could learn from? Another way to look at this is blockchain is growing at 23% YOY over the last 13 years. What technologies should we compare it to?
Just to pull on the thread, is blockchain more like "hard tech"? If so, that would be really odd since it's not "hard tech". What has EV growth been like?
Here's a chart [1] from [2] that says about 17 million EVs have been sold since 2012. Let's call that 13 years for the sake of keeping the math the same, but the web grew 100x as fast. OK so cryptocurrency is growing about order of magnitude faster than EVs, but why is it slower than the web? What else develops at a speed that's 10x faster than the growth of EVs, but still slower than the web itself?
I'm actually working on a tool to use data from the Glassnode API to possibly answer some of the questions. I'll be pushing it to Github at some point in the near future.
What's interesting about public blockchain data (from a node directly or something like Glassnode) is that the transaction data is so readily available. We can easily analyze transactions and unique addresses. Therefore transaction rates and volumes, long with unique to/from addresses, allow for ready analysis of adoption across the entire chain. For example, Polygon has relatively low trade volume but is very popular for smart contracts (applications). Because of this the ATH high daily transaction count is roughly 9x Ethereum even though Ethereum is considered to be more "popular".
I don't have any better or more valid comparisons for blockchain than what I've presented in that rant. Certainly open to any suggestions. That said, I think there's probably a valid internet law that can be derived stating something like "A new technology paradigm either reaches 1 billion users inside of 13 years or becomes irrelevant" (with the web setting the benchmark at 1B in 13 yr).
Appreciate the feedback on image and title. Perhaps a little clickbaity but I feel the premise is original and I establish some bonafides immediately.
The web and mobile were absolutely hard tech. The majority of web users went from dialup ISPs to cable, DSL, etc. Many thousands of truck rolls, ditches dug, submarine cables laid, increasingly scalable servers, the C10k problem[1], fiber optical cabling and optical switching innovation, etc. The list is almost endless.
Mobile - remember upon the exclusive release of the iPhone on AT&T and the network practically collapsed in major cities[2]? The success of mobile depended on massive build out of additional cellular infrastructure, back hauls, new spectrum auctions, the sun-setting of analog TV broadcasts, etc. Look at the standards (and component standards) for 2G-5G. I'd say that alone makes it hard tech.
EVs are a really hard (potentially invalid) comparison if for no other reason than the average car on the road in the US is 12.1 years old[3]. I don't know how I would start looking at EV adoption for comparison with that thrown in the mix.
Cool! You've probably already seen it but Dune Analytics has some useful blockchain data analysis tools too. I hadn't heard of Glassnode before.
Edit: thinking of other technology adoption curves. Faster than EVs but slower than web and mobile. Thinking of finance too. What's the adoption of NFCs/touchless payments/chips look like? What about the rise of microlending/lending club etc?
Looking at other technology adoption curves is interesting and difficult. Bitcoin (the "digital cash" vision) was released in 2009. It didn't register on Google Trends until 2011. This already isn't a good sign.
Ethereum came along in 2015 and introduced the concept of the Ethereum Virtual Machine and smart contracts. Now "blockchain" is a bona fide platform where people can leverage Ethereum and the underlying blockchain network to build things. Another question - why did it take six years for blockchain technology to progress from digital cash to platform?
Now seven years after that even when combining digital cash (payments), "store of value"/gambling (trading, speculation), and all things smart contract (NFTs, DAO, ENS, microlending, etc) there are still at best 100 million total users for all of these use cases combined.
CERN likes to say April 30, 1993[1]. The web was the first "killer platform" (built on top of the internet) that enabled virtually all killer apps to come - to the point where the general population uses the word "internet" to refer to the web (depending on how you view mobile apps).
The web is a platform that stands on top of the internet (which is at OSI model layers 1-4).
Blockchain is a platform that stands on top of the internet too. Difference is that after 13 years no one has been able to create anything of utility with it other than trade tokens.
There's no point in bringing up the history of the plumbing that is the internet from before either of these dates. Besides, as other commenters have noted, the realities of connectivity in the early nineties are so radically different from the release of Bitcoin in 2009 it's absurd to compare the two.
Yet, even when you do, the web still grew at 10x the rate of blockchain in terms of user adoption.
I hear all your data points. If cryptocurrency fails to take off 10 years from now, we can look back and say "oh we should have known back in 2021". But what do these data points mean if cryptocurrencies take off and change the world over the next decade?
As per usual, someone loosely understands one part of blockchain and - using the craziness that is "Web 3" at the moment - attacks the entire technology.
> The cryptocurrency web3 starts with all our existing infrastructure. So I still need a DNS name, I still need a server, I still need storage, and I still have a distributed computation occurring between the browser and the server.
No, you could work off of static IP addresses or indeed decentralized naming services like ENS, use smart contracts, use IPFS, but I agree you do still need to use the internet.
> When a user wants to use my service they will pay some amount of cryptocurrency to perform the cryptocurrency-side computation, with any remaining transferred to me as a fee for my service.
This is one case for one business model yes, but none of these are required.
> Put bluntly, the Ethereum “world computer” has roughly 1/5,000 of the compute power of a Raspberry Pi 4!
Yes, if we completely ignore recent developments in Layer 2 technologies and chains.
> What about the storage? The entire Ethereum blockchain is just 1 terabyte of data and adds a total of a few hundred kilobytes a minute. Storing the Ethereum blockchain using a robust commercial service like Amazon S3 costs just $20 a month.
Yes, if we completely ignore recent developments in off-chain storage or Layer 2 technologies and chains. Also, different use cases - I don't think anyone sold Ethereum as a replacement to storage.
> I know some cryptocurrency enthusiasts will protest that their favorite blockchain is cheaper than Ethereum. And it’s true, underutilized cryptocurrencies may be one or two orders of magnitude less expensive to use in the “web 3” vision when compared with Ethereum.
Nope, this difference in fees is down to other factors like PoW vs PoS, a topic the author didn't mention once (and might not actually know about).
> So why this hype? Because the cryptocurrency space, at heart, is simply a giant ponzi scheme where the only way early participants make money is if there are further suckers entering the space. The only “utility” for a cryptocurrency (outside criminal transactions and financial frauds) is what someone else will pay for it and anything to pretend a possible real-word utility exists to help find new suckers.
No, this is true for some of the public cases you see mentioned 50 times a day and I understand why people think this but it isn't true. One obvious area where it isn't some Ponzi scheme is the developments in inter-bank trading, such as https://en.wikipedia.org/wiki/Central_bank_digital_currency. Goldman Sachs has been trading Digital Treasury Bonds with JP Morgan on JPM's Onyx blockchain, Visa is working on an international payment system using a blockchain, Swift is building projects on blockchains https://www.swift.com/your-needs/data-and-technology/distrib..., etc...
> After all, a programmer doing the most basic test of a web3 prototype is going to need to get the cryptocurrency, spend the cryptocurrency, and any application will require all users to get the cryptocurrency as well.
A programmer would need testnet currency to test it and then a couple dollars to deploy it on mainnet. Some apps don't require users to use cryptocurrency. FIDE has an official online chess app built on Algorand that you wouldn't realize was a blockchain unless you know who Algorand are.
> If this gets abandoned quickly due to the inevitable technical failure “web3” still accomplished its goal of getting more suckers in and extracting their money.
"Web 3" is a term used more by media and people complaining than the people building products but, as someone who's been in the blockchain space for years, can promise has never once had to work with someone like this. I'm sure they exist - you see crap projects every day that are scams and yes it's a murky space because of this - but labelling everyone with this brush is unfair to the rest of us. If you want to put a moral hat on, maybe start looking at the damage being done in all the other areas of tech too, such as addictive applications, social medias impact on democracy and mental health, etc...
> So in the end web3 is a con job, a technological edifice that is beyond useless as anyone who attempts to deploy a real application will quickly discover.
Real projects in the wild doing real things mentioned above - sadly interbank trading of digital treasury bonds is just pretty boring so aren't as publicized as a 1000x JPG someone laundered. Just please stop assuming that covers everyone and everything in the blockchain space.
The problem with all of the commercial blockchains you mentioned is that their security properties are more akin to having a shared Git repository with the ledger data in it rather than Bitcoin. The fundamental property that people think of when they say "blockchain" is that all of the ledger data can be validated by any suitably-equipped network participant. This 100% trustless design is interesting, but the security model it relies upon prohibits pretty much anything more than tracking tokens issued entirely on-chain.
The moment you want a blockchain to interact with the real world, you need to introduce trust back into the system. Which removes much of the claimed benefit of blockchain. Such systems might still be useful to people, of course; but that's not what the blockchain/Web3 people actually want to sell. The elevator pitch of blockchain is "trustless decentralized finance", and Web3's is "replace advertising with financialization as the funding mechanism for online media". Both concepts do not hold up to the marketing hype, even if there are use cases for them.
> Yes, if we completely ignore recent developments in Layer 2 technologies and chains.
Yes I’d prefer to ignore it and save myself the trouble of sifting through the endless avalanche of crypto marketing for novel ideas that solve a problem other than “line go up”.
It can’t just be enough to write a conventional program that runs on a computer to help solve a problem. No, rent must be charged.
This article came up on twitter last month. I remember finding it very interesting and taking the time to systematically untangle it. I will try to do a good job of explaining all the problems here.
This article is interesting to analyze regardless.
First; the author is correct across several dimensions. The Ethereum virtual machine is non optimal as a computer (currently). Blockchain is also not a great way to store data (currently). Let’s agree with the author on these points.
However, no one is saying anywhere that Ethereum is meant to replace AWS or that it’s primary purpose is to replace AWS S3. The nomenclature of “world computer” is not currently meant to be understood in the sense of “AWS killer.”
Ethereum was not designed to be a super fast computer to calculate things. It was designed to decentralized money and enable smart contracts.
Right now, Ethereum and crypto primarily replacing money, first and foremost. There are numerous people working on subsequent innovations called “Decentralized Web Services” which WILL have the intent to compete with AWS. So this angle of attack on Ethereum is unjustified.
It makes sense to me that all compute resources are ultimately listed on an open marketplace where people can bid for them. That structure might compete with AWS. I don’t think that is the first thing Ethereum is designed to do, but Ethereum and technologies like it will enable that use case later.
Author goes in to leap to a conclusion which can’t be justified Aka Web3 is all a scam. Wrong. Every problem he identifies is well understood by leading folks and in the process of being solved across multiple dimensions. Just read Vitalik’s blog. Vitalik has talked about solutions for years.
Author just points out a couple known problems in Ethereum that are being solved actively and goes “see, it’s all a scam,” not a good angle.
So intent of author feels disingenuous. Or lazy?
Author also points out that Web3 relies extensively on existing internet infrastructure and then makes the conclusion that “it’s not decentralized and has gatekeepers.”
He calls reliance on DNS, browsers, mobile phones and AWS as being “gatekeepers.” The reality is that those things are truly only “gates” if you can’t transition away from them or substitute them immediately.
Aka: Even if you have some aspects of your infrastructure which rely on “web2,” if you can substitute them for alternatives s then one could point out that there is no fundamental blocker to building decentralized systems on top of that.
My view is that crypto / Web3 are internet phenomena and internet technologies. Not replacements for the internet. So obviously they need to use internet stuff like DNS.
I have heard people talking about building private internets before, but I don’t think anyone in Web3 / crypto are claiming that they are going to replace the internet anytime soon. There are also many discussions around how to decentralize databases via IPFS, and replace DNS with something else.
Compute marketplaces such as GridCoin, Akash Network, etc are under development.
So those “gatekeepers” are getting eroded in real time, but browser and phone likely to remain.
The virtual machine is also under “improvement.” Folks are actively innovating new approaches to application specific VMs which will speed things up for specific use cases.
Furthermore, sovereign blockchains and subnets will allow much higher performance use cases for Web3 at the edge. Put all this together and most of the criticisms fall away.
My personal opinion is that we are going to see all tech stacks rebased around wallets as the core construct and this is going to force a rewrite of nearly everything.
> Ethereum was not designed to be a super fast computer to calculate things. It was designed to decentralized money and enable smart contracts.
Those are counter to each other. Money transactions are fast and free. Crypto transactions are sow and expensive. To handle being money or contracts, it has to be a super fast computer
https://news.ycombinator.com/item?id=29587469