Moxie Marlinspike (pictured), the founder of Signal chatapp with its 40 million monthly active users and previously head of security at Twitter, has stated “nerds like me are excited to build for” web3.
In the latest sign of the growing prominence that web3 is attracting among coders, Marlinspike further said:
“Web3 is, at the very least, something new on the nerd level – and that creates a space for creativity/exploration that is somewhat reminiscent of early internet days.”
He does not think that web3 is “on a trajectory to deliver us from centralized platforms” however, stating:
“I’m hopeful that the creativity and exploration we’re seeing will have positive outcomes, but I’m not sure if it’s enough to prevent all the same dynamics of the internet from unfolding again.”
Marlinspike instead argues that “there is nothing particularly ‘distributed’ about the apps themselves: they’re just normal react websites. The ‘distributedness’ refers to where the state and the logic/permissions for updating the state lives: on the blockchain instead of in a ‘centralized’ database.”
He claims that the communication to that blockchain is mostly through Infura or Alchemy, which led to a situation where an NFT that shows a different image to different people would not show up on his wallet after OpenSea removed it.
OpenSea has not yet replied to our request for clarification with it unclear why the NFT was removed as Marlinspike says he was upfront about this NFT in effect ‘hacking’ the URL pointer to show a different image to different people.
Nor is it clear what wallet he was using exactly that would no longer show his NFT, with Marlinspike claiming the OpenSea API is being used by wallets to show NFTs instead of the blockchain/ipfs.
He does not mention IPFS however, claiming instead NFTs “contain a URL that points to the data… there’s no hash commitment for the data located at the URL.”
The ipfs url is a hash of the content, and there are quite a few project now where the nft itself is on the blockchain through the algorithm being put down on the smart contract, allowing you to re-create the art work.
Marlinspike however seems to be a bit new to this space and sounds unfamiliar with a lot of important details that one isn’t able to easily come across when engaging in a surface hands on as was the case here with Marlinspike so creating an NFT and trying to create a derivatives market for it.
Good luck with the latter, it might be interesting, but it isn’t very clear whether he’ll actually bother because he seems inflicted with the ‘can’t see anything here’ syndrome in effectively suggesting this web3 is actually web2 and the blockchain part will eventually be removed with Opensea somehow becoming just a jpeg market:
“The people at the end of the line who are flipping NFTs do not fundamentally care about distributed trust models or payment mechanics, but they care about where the money is. So the money draws people into OpenSea, they improve the experience by building a platform that iterates on the underlying web3 protocols in web2 space, they eventually offer the ability to ‘mint’ NFTs through OpenSea itself instead of through your own smart contract, and eventually this all opens the door for Coinbase to offer access to the validated NFT market with their own platform via your debit card…
Eventually, all the web3 parts are gone, and you have a website for buying and selling JPEGS with your debit card. The project can’t start as a web2 platform because of the market dynamics, but the same market dynamics and the fundamental forces of centralization will likely drive it to end up there.”
With such pessimists building web2, no wonder our internet has devolved into a Septembered mess. Marlinspike for his part is thought to be behind Mobilecoin, which itself looks a bit of a mess on the surface, with Signal integrating that Mobilecoin in UK according to announced plans this summer.
For some reason it looks like web2 founders tend to prefer launching their own blockchain, instead of developing on an established one, but the rise of web3 may change that with a big debate seemingly going on, a debate that maybe is turning in our favor.
The Centralized Illusion
In his critique Marlinspike effectively suggests that just as web2 has Google, Facebook, and maybe Signal, so too web3 has Opensea, Infura and Metamask.
He makes two main points, both in our view misleading. The first one is an implicit suggestion that ethereum has just two nodes, Infura or Alchemy. He emphasizes that by repeatedly stating people do not want to run their own servers, and a node he says is a server as it runs on a server where a website is concerned.
We have some data on this point. Bitcoin started off with some 50,000 nodes at one point or even more, with it going down to eventually about 5,000 nodes in 2016.
As it happens ethereum also had 50,000 node or more in 2016, and currently has 5,400 nodes with 2021 for eth being bitcoin’s 2017 in our view.
Bitcoin has tripled its numbers since, with it currently having 14,215 nodes. This fits fine with our theory that as users grow, node numbers should too because there will be more businesses, entities or individuals that want or need to run a node.
Someone like Peter Todd would and has before disputed this theory because of things like Infura, a node as a service startup. The data seems to back us however, with Infura being a convenience but potentially at a cost as you’re outsourcing validation and it is useless if you’re mining or staking.
Metamask uses them as default, with both Metamask and Infura being ConsenSys incubated, but you can point Metamask to your own node if you want or some other node.
Many don’t bother, and that’s in great part because of Simple Payment Verification (SPV) wallets. Now someone like Todd would say these aren’t ‘real,’ we never managed to create SPVs as contained in the bitcoin whitepaper, but most consider the wallets to have good enough cryptography methods to prevent the node from lying.
Not perfect and we’re not going to go into the details because a cryptographer like Marlinspike should look them up himself, but in many years now there haven’t been any problems with SPV wallets which generally work pretty fine.
Metamask happens to be such wallet, and like all SPV wallets they need a node to connect to. There can potentially be improvements here as they can potentially connect to many nodes, but Infura is their bros so if you don’t like the default, then change it.
Marlinspike points out that Metamask also relies on etherscan’s API, his implicit and unstated point being that Infura and etherscan are centralized platforms and so kind of web2.
They’re nodes however, not platforms. They can not lie without pre-planning and targeting of an SPV wallet, and if they did. it’s good for bitcoin because it would make those SPV wallets even more robust by effectively making multi node connections practically mandatory.
The effort to lie would in addition be great in that a fake block has to be created. For a user transfer, this might be more just an inconvenience. You thought you sent the funds, but Coinbase says you haven’t and eventually you might find out that although you thought you had made a transfer, you were fooled because they can only manipulate the data to potentially fool into a double spend, they can’t steal your funds or change ownership otherwise.
This has never happened as far as we are aware except some very rare more generalized double spend by some rogue mining pool employee, but it is possible and for something like Coinbase it may even be worth it, which is why they run their own nodes and not SPV wallets or even Infura.
Now much of course depends on execution. The specifics for Metamask might be different, we haven’t looked since 2016 when it came out, but conceptually the point is that the node can’t easily lie, and even if they did it would be to double spend which would mean fooling a receiver, someone like Coinbase, which is why Coinbase should and does full node.
Very simple stuff, but to be fair to Marlinspike everyone has gone through the same considerations, questions, and debates during their own ‘initialization’ so to speak, and thus what we’d call extreme simplification in understanding, is understandable.
That said Todd would agree with him, but he hasn’t come up with a more decentralized solution so it’s like shouting at clouds really when people do have the option to run full nodes or connect to their own node, and if they don’t, they don’t. Some will because they need to and that is enough to keep the network decentralized as well as to keep the option of running your own node whenever you want.
Which is why “most participants don’t even know or care it’s happening.” We do know, and he’s right that we don’t care because what exactly is there to care about or is he suggesting some puritanism where we compel the mandatory running of a personal node?
Some might care, Todd does or at least he talks it, but we were right and he wasn’t. Bitcoin’s node numbers are up. We’ll see if eth’s numbers will also increase in the next four years presuming demand grows. You’d think they would not least because with staking you make nodes even more mandatory than just for verification.
The final point on this matter is that this issue is systemic not individualistic. It isn’t our business conceptually speaking what Metamask chooses, but it is our business if you can run your own node and it probably would have been our business too at the beginning if Metamask did not give you the option to connect to your own node.
Marlinspike however is probably intending to argue that people no wanna run nodes, they’ll infura, and thus there is no network. It’s Todd’s argument. An old one and mistaken because well people are running nodes and in bitcoin there’s now 3x more than four years ago. In eth there will too, even more than that perhaps because of staking.
Opensea. We would be very interested to know what exactly happened in that NFT that disappeared from wallets as well.
The story of NFTs from where we stand is Cryptokitties for 5 minutes four years ago, then NFT-ed insurance in 2020 gradually blew up into what is now seemingly being perceived as centered on Opensea.
Our Metamask has shown Cryptokitties for ages. Not really any other NFTs. So what wallet was he using exactly? Maybe he is referring to his own wallet on Opensea, his own profile page. It may well be some wallets use Opensea’s API, but Opensea just uses Google for caching and they do have competitors like Rarible.
The conceptual point to make here however is that popularity is not centralization. Centralization is the lack of meaningful choice, while after years of discussions and analysis, we have concluded that the definition of decentralization is forkeability.
Whether we like or dislike Opensea is far less relevant than whether Opensea has anything we can’t replicate. Since the accounts themselves and all the data is on the public blockchain, they do not.
We can’t fork Signal. Maybe we can copy its public data, but not its accounts, its usernames and passwords. We can scrap Facebook, but our grandma can’t log into newFacebook because we don’t have her username and password. We can fork Opensea, in fact soon enough there may even be templates that in time we can buy for $20 and now anyone can connect their Metamask to newOpensea and be no different than on oldOpensea.
Marlinspike points out that Opensea has added commission for every NFT resale, something the NFT standard lacks. That can be added to ERC-721 as an option and would easily be replicable in our newOpensea anyway since again everything is happening on the blockchain.
Now you take out that blockchain as Marlinspike suggests, and you have nothing because people aren’t buying jpegs, they’re buying tokens, or to someone new like Marlinspike, they buying what they think are bitcoins which happen to have pretty pictures.
Without the tokens, without the open ethereum accounts, without the estimatebly long lived storage in ipfs, and if not long lived then the best option at it currently, there is no NFT and there is no Opensea.
The skepticism thus is a bit too much, or maybe we getting a bit tired of it a decade on now, but in this case it’s because web3 appears to be at a pivotal point.
The Battle For Coders
“There is a giant sucking sound coming from crypto,” said Sridhar Ramaswamy, chief executive of search engine start-up Neeva and a former Google executive, who competes with crypto companies for talent. “It feels a bit like the 1990s and the birth of the internet all over again. It’s that early, that chaotic and that much full of opportunity.”
So the public is told under the headline “Apple among tech companies quickly losing talent to lucrative crypto startups.”
The examples are too many, with this starting in 2018 when pay packages “gotten insane.” At the same time the seething in some tech corners seems to have increased with the same intensity.
Poor Firefox was made to back down from accepting crypto donations, free money. Discord is now ripe for disruption after likewise backing down.
In r/tech-anything you now only hear ree ree ree when it comes to cryptos. Scam, ponzi scheme, or the more polite version of “gold rush” from Marlinspike is all you hear. Same old rubbish to which Buttcoin has desensitized us since 2014, not least because we’ve had plenty of bears since then to Buttcoin ourselves.
If some think we’re deluded, or there’s nothing here, or the favorite ‘a solution looking for a problem,’ they should have looked in 2015 when we had a whole civil war, or in 2019 when the tiniest little fault was screeched out.
We’re not deluded, bears made very sure of that. We’re veteran survivors of a mighty revolution that isn’t utopia but is better in some ways.
It has problems of course, and some looking at Marlinspike’s piece may only see problems and dismiss all of it. But some, who may be smarter, especially if they’re coders or a startup, may well think that problems means a lot of money as providing a solution to a problem is precisely how you make them.
And it is those opportunities which appears to have so uniquely attracted coder’s attention. Web3 in particular appears to have grabbed their imagination. It’s a curious narrative and even vision that can’t easily be articulated because the details are yet to be created.
If we articulated it all in one word, it would be freedom. Not necessarily real because we don’t know yet, but a shot at it.
The freedom for coders to once more be at the forefront. For garage kids to once again rise to empire. For status or residence to be irrelevant, whether in Calcutta or at the Queen’s palace.
The freedom to tear down the Securities Act 1933, and ironically the lefties are against us when this protects the very rich by the iron force of unjust law. But however much they may screech, we will fight for it and win because they can do nothing to our code network.
The thing about freedom is that some hate it. The events in Kazakhstan show just how much. The likes of Marlinspike, though we know pretty much nothing about him and intend not in any way to attribute to him what follows rather than generally, probably would like very much coders in 9-5 boxes rather than garage empiring.
Yes, it’s not utopia. Yes the websites are react. Yes it’s not quite a completely alien world this web3. But it’s a new world. Our world. The student’s revolution of 2013.
We might even get no where, but what else is there. Stocks that start at $100 billion market cap, some 4x the GDP of El Salvador, a whole country? What are you gone do with gold? There is some innovation in
At least here for now, we have the power. Maybe we’ll lose it one day. Maybe corporate boards will take over again. Maybe maintaining forkability will be too difficult. But we haven’t lost it yet, and we do fork for breakfast.
It is also about who do you want your talent to enrich. Yourself of course, but is it also the students – considering this space was open to all since inception and thus has plenty of ‘normies’ so to speak since inception or near there – or the already powerful?
And to those powerful, like Marlinspike, this isn’t a French revolution. We admire plenty of the rich and powerful, but we admire and even adore competition and meritocracy far, far more.
They’re welcomed. Some ree at VCs. We have no problem with VCs. We have a big problem with the Securities Act 1933 that prohibits competition against them.
We have no problems with the players, or even the game, we just think we have a better one. Our fathers and grandfathers and where it comes to web2 our older brothers, have set up systems under the conditions of their time and technology. We happen to have a new one, and thus from stocks you don’t actually own or private ‘public data,’ they’re not ‘wrong,’ they’re just worse.
They’re not wrong because there wasn’t another option. The stock or paper system is flawed from where we stand, and the control of our input by a handful is also very open to abuse, but at least where the paper system is concerned our fathers have tried their best to make it as just as they possibly can.
That is to say we do not think there are evil manipulative forces trying to milk cows, but that their technology and capabilities were and are inferior to our new capabilities.
All very simple. Which is maybe why that ree ree ree falls on deaf ears as the smarter ones are clearly able to see, and those are the ones we need right now so that this new world can be refined for the Septembered ones.