My First Impressions of Web3

By moxie

While I consider myself a cryptographer, I’m not particularly interested in “crypto.” I can’t remember if I ever actually said “get out of my way,” but instead of clicking on the latest NFT release, I’m more likely to scroll through some nostalgic “Pepperidge Farm Remembers”-style memes about the good old days when “crypto” meant “cryptography.”

Also, frankly, I don’t buy into the generational excitement of moving every aspect of life to a tooled economy.

Even on a technical level, I’m not yet a believer. So, given all the recent attention paid to so-called Web3, I decided to explore some of the developments in this space in more depth to see if I had missed anything.

My take on Web1 and Web2

Web3 is a relatively vague term, which makes it difficult to rigorously evaluate the goals of Web3. But the general view (https://a16zcrypto.com/posts/article/why-web3-matters/seems to be thatWeb1 was decentralized, Web2 centralized everything onto platforms, and Web3 will be decentralized again. Web3 should be able to give us the same rich experience as Web2 on a decentralized basis.

First, let's clarify why centralized platforms are a good thing, and in my opinion the explanation is very simple:

1. People don't want to run their own servers, and never will. The premise of Web1 is that everyone on the Internet is a producer and consumer of content, and a producer and consumer of infrastructure. We all have our own web server, our own website, our own mail server, our own status message server, our own chargen server, our own character generation server. However, I don't think this can be stressed enough: This is not what people want. People don't want to run their own servers. Even geeks don't want to run their own servers right now. Even organizations that develop software full-time don't want to run their own servers right now. If there's one thing that sums up our understanding of the world, it's this: People don't want to run their own servers. Companies that offer server hosting services have succeeded because of this, and companies that have iterated on new features based on the possibilities of these networks have been even more successful. 2. Protocols evolve much slower than platforms (https://signal.org/blog/the-ecosystem-is-moving/). After more than 30 years, email is still unencrypted; meanwhile, WhatsApp went from unencrypted to fully end-to-end encrypted in a year. People are still trying to standardize how to reliably share videos over IRC; Slack lets you create custom reaction emojis based on your face. This isn’t a funding issue. If something is truly decentralized, it becomes very difficult to change it and tends to stagnate. This is a problem for technology because the rest of the ecosystem is moving fast, and if it can’t keep up, it will fail. There’s a whole parallel industry focused on defining and improving methodologies like Agile, trying to figure out how to organize large teams to move as fast as possible, because it matters.

This is a problem when technology itself tends to stagnate rather than evolve. One proven way to do this is to centralize a protocol that’s stuck in the 90s and then iterate quickly.

But Web3 wants to be different, let’s take a look. To quickly get a feel for the space and better understand possible future developments, I decided to build a few dApps and create an NFT.

Make some decentralized apps

To get a feel for the Web3 world, I made a dApp called Autonomous Art (https://autonomous.graphics/) where anyone can mint tokens by making visual contributions to NFTs. Over time, the cost of making visual contributions increases, and the minting funds paid by contributors are distributed to all previous artists (visualize this financial structure as a pyramid-like shape). At the time of writing, over $38,000 has been invested to create this collective art work.

I also made a dApp called First Derivative (https://firstderivative.market/) which allows you to create, discover, and exchange NFT derivatives that track an underlying NFT, similar to how financial derivatives track an underlying asset?.

Both gave me a feel for how this space works. To be clear, there’s nothing particularly “distributed” about these apps themselves: they’re just regular React websites. 'Distributed' refers to where the state and the logic/authority to update it resides: on the blockchain, not in a 'centralized' database. One thing that strikes me as odd in crypto is the lack of focus on the client/server interface. When people talk about blockchains, they talk about distributed trust, leaderless consensus, and all the mechanics of how it all works, but often gloss over the reality that clients ultimately have no part in those mechanics. All of the network graph is of servers, the trust model is between servers, and everything is about servers. Blockchains are designed to be a peer-to-peer network, but they are not designed so that your mobile device or browser can actually be one of those peer-to-peer networks.

With the paradigm shift to mobile, we now live in a world of clients and servers, with the former being completely incapable of replacing the latter: these questions are more important to me than ever. Meanwhile, Ethereum actually refers to servers as “clients” , thus not even having a word to describe the actual, trustless client/server interface that must exist, and not realizing that if successful, there will ultimately be billions (!) more clients than servers.

For example, dApps like Autonomous Artor

These client APIs don’t use anything to verify the authenticity of the blockchain state or responses. The results aren’t even signed. An application like Autonomous Art would say “hey, what is the output of this view function on this smart contract”, and Alchemy or Infura would respond with a JSON blob saying “here is the output”, and the application would render it.

This is amazing to me. People have invested a lot of work, energy, and time into creating a trustless distributed consensus mechanism, but almost all clients that want to access it simply trust the output of these two companies without any further verification. This doesn’t seem like the best privacy situation either. Imagine that every time you interact with a website in Chrome, your request is sent to Google before being routed to its destination and back. This is the situation on Ethereum today. All write traffic is obviously already public on the blockchain, but these companies can also view almost all read requests from almost all users in almost all dApps. Blockchain supporters may say that if these types of centralized platforms emerge, it’s fine because the state is viewable on the blockchain, so if these platforms misbehave, customers can simply move elsewhere. However, I think this is a very superficial view of the dynamics of platform development. Let me give you an example.

Making NFTs

I also wanted to create a more traditional NFT. Most people think of images and digital art when they think of NFTs, but NFTs don't typically store data on-chain. This would be too expensive for most NFTs of most images.

NFTs do not store data on-chain, but rather contain a URL pointing to the data. What surprised me about the standard is that there is no hash commitment to the data at the URL . Looking at many of the NFTs sold on popular markets for tens, hundreds, or millions of dollars, the URL is often just pointing to a VPS somewhere running Apache. Anyone with access to the machine, anyone who buys the domain in the future, or anyone who breaks into the machine can, at any time, change the image, title, description, etc. of the NFT to whatever they want (regardless of whether they "own" the token or not). There is nothing in the NFT spec that tells you what the image "should" be, or even allows you to confirm that something is the "correct" image.

So as an experiment, I made an NFT that changes based on who is viewing it, because a web server serving an image can choose to serve a different image based on the requester's IP or user agent. For example, it looks one way on OpenSea and another on Rarible, but it will always show up as a big ? emoji when you buy it from a crypto wallet and view it. What you bid is not what you get. There is nothing unusual about this NFT, it's just the way the NFT spec is built. Many of the highest priced NFTs could turn into ? emojis at any time; I just made that explicit. A few days later, without any warning or explanation, the NFT I made was removed from OpenSea, an NFT marketplace: The removal indicated that I had violated some terms of service, but after reading the terms, I did not see any clause prohibiting NFTs that change based on where they are viewed, and I publicly described it as such.

What I found most interesting, though, was that after OpenSea deleted my NFT, it also no longer appeared in any crypto wallets on my devices . But, this is Web3, how is this possible?

Crypto wallets like MetaMask, Rainbow, etc. are “non-custodial” (keys are kept client-side), but have the same problem as my dApp above: the wallet must be run on a mobile device or in a browser. Meanwhile, Ethereum and other blockchains were designed with the idea that it was a peer-to-peer network, but were not designed so that your mobile device or browser could actually be one of those peer-to-peer networks. Wallets like MetaMask need to do basic things like show your balance, recent transactions, and NFTs, as well as more complex things like constructing transactions, interacting with smart contracts, etc. In short, MetaMask needs to interact with the blockchain, but the blockchain is built in a way that a client like MetaMask cannot interact with it. So, like my dApp, MetaMask does this by making API calls to three companies that have integrated in this space.

For example, MetaMask shows your recent transactions by making an API call to etherscan:

GET https://api.etherscan.io/api?module=account&address=0x0208376c899fdaEbA530570c008C4323803AA9E8&offset=40&order=desc&action=txlist&tag=latest&page=1 HTTP/2.0 

…by making an API call to Infura Call to display your account balance:

POST https://mainnet.infura.io/v3/d039103314584a379e33c21fbe89b6cb HTTP/2.0{    "id": 2628746552039525,    "jsonrpc": "2.0",    "meth od": "eth_getBalance",    "params": [        "0x0208376c899fd aEbA530570c008C4323803AA9E8",        "latest"    ]}

MetaMask doesn't actually do much, it's just a view into the data provided by these centralized APIs. This isn't a MetaMask-specific problem: do they have alternatives? Rainbow et al are set up in exactly the same way. (Interestingly, Rainbow owns their own data for the social features they’re building in the wallet: social graph, showcase, etc., and chose to build all of this on Firebase rather than a blockchain.)

All this means is that if your NFT is removed from OpenSea, it will also disappear from your wallet. Functionally, it didn’t matter whether my NFT existed permanently on the blockchain somewhere, because wallets (and increasingly other things in the ecosystem) simply used the OpenSea API to display NFTs and for queries for NFTs owned by my address, it started returning 304 No Content !

Reinventing the World

Given the history of Web1 becoming Web2,what’s odd about Web3 to me is that technologies like Ethereum are being built with many of the same implicit features as Web1.to make these technologies usable, the space is consolidating around platforms. Again. People will run servers for you and iterate on new features as they emerge. Infura, OpenSea, Coinbase, Etherscan.

Similarly, Web3 protocols are moving slowly. When building First Derivative, it would have been nice to be able to set the minting price of a derivative as a percentage of the underlying value. That data isn't on-chain, but it's in the API that OpenSea will provide to you. People are excited about NFT royalties because they benefit creators, but royalties aren't specified in ERC-721 and it's too late to change that now, so OpenSea has its own way of configuring royalties that exists in the Web2 space. Rapid iteration on centralized platforms has moved beyond distributed protocols and integrated control into the platform.

Given these dynamics, I think it’s not surprising that we’re already in a situation where your crypto wallet’s view on NFTs is what OpenSea’s view on NFTs is. I don’t think we should be surprised that OpenSea isn’t a pure “view” that can be replaced because it’s been busy iterating the platform beyond what can be achieved by strictly adhering to standards that are impossible/difficult to change.

I think this is very similar to the situation with email. I can run my own mail server, but functionally it doesn’t matter for privacy, censorship resistance, or control because GMail will be on the other end of every email I send or receive anyway. Once a distributed ecosystem centralizes around one platform for convenience, it becomes the best of both worlds: centralized control, but still decentralized enough to falter over time. I can build my own NFT marketplace, but it doesn’t provide any additional control if OpenSea mediates the view of all NFTs in the wallets people use (and all the other apps in the ecosystem).

This is not a complaint against OpenSea, or a rebuke of what they’ve built. Quite the opposite, they’re trying to build something that works. I think we should expect this kind of platform consolidation to happen, and given the inevitability of it, we should design systems that give us the functionality we need when organized in this way. However, my sense and concern is that the Web3 community expects something different than what we’re seeing so far.

It’s still early

“It’s still early” is the most common statement I see in the Web3 space, especially when discussing issues like this. In a sense, the failure of cryptocurrencies to get beyond a relatively preliminary engineering phase is exactly what allows people to consider this period “early”, since objectively it’s been around for a decade or more.

However, even if this is just the beginning (and it very well may be!), I’m not sure we should take that as comfort. I think the opposite may be true; it seems we should note that, from the outset, these technologies tend to be centralized through the platforms that enable them, with little negative impact on the velocity of the ecosystem, and most participants don't even know or care that it's happening. This may suggest that decentralization itself is not actually of immediate practical or urgency to most people downstream, that the only level of decentralization people want is the minimum required for something to exist, and that without very conscious consideration of this, these forces will over time move us further away from the desired outcome, not closer to it.

But You Can’t Stop the Gold Rush

If you think about it, OpenSea would actually be a lot “better” if all the Web3 parts went away. It would be faster, cheaper, and easier to use. For example, to accept a bid on my NFT, I would have to pay over $80-$150+ in Ethereum gas fees. This sets an artificial threshold for all bids to be accepted, otherwise you would lose money by accepting a bid below the gas fee. Credit card payment fees often feel expensive, but are cheap compared to this. OpenSea could even publish a simple transparent log if people wanted to publicly record trades, offers, bids, etc. to verify their accounting.

However, if they built a platform for buying and selling images that was not nominally based on cryptocurrency, I don’t think it would catch on. Not because it’s not distributed, because we’ve seen that a lot of the things that are needed to make it work are not distributed already. I don’t think it would catch on, because it’s a gold rush . People are making money speculating in cryptocurrencies, and those people are interested in spending those cryptocurrencies in a way that supports their investments while providing additional returns, so that defines the context of the wealth transfer market.

The people at the end of the line, flipping NFTs, don’t fundamentally care about distributed trust models or payment mechanisms, but they care about where the money is. So money attracted people to OpenSea, they improved the experience by building a platform that iterated on the underlying Web3 protocol in the Web2 space, they ultimately provided the ability to “mint” NFTs through OpenSea itself rather than through your own smart contract, and ultimately this all opened the door for Coinbase to provide access to a verified NFT market through their own platform through your debit card. This opened the door for Coinbase to manage the tokens themselves through dark pools held by Coinbase, which helped eliminate transaction fees and made it possible to avoid interacting with smart contracts entirely. Eventually, all the Web3 parts go away, and you have a site where you buy and sell JPEGS with a debit card. The project couldn't start as a Web2 platform because of market dynamics, but the same market dynamics and fundamental forces of centralization will likely drive it there eventually.

At the end of the stack, NFT artists are excited about this progress because it means more speculation/investment in their art, but if the purpose of Web3 is to avoid the pitfalls of Web2 then we should be concerned that this is already the natural trend of these new protocols that are supposed to offer a different future.

I think these market forces are likely to continue, and the question of how long it lasts, in my opinion, is whether the massive accumulation of crypto ends up being inside the engine or in a leaky bucket. If the money flowing through NFTs ends up being directed back into the crypto space, it could accelerate forever (regardless of whether it's just Web2x2 or not). If it pours out in large quantities, then it will be a blip. Personally, I think there's enough money being made and enough faucets to keep it going that it won't be just a blip. If that's the case, it seems worth considering how to avoid Web3 becoming Web2x2 (Web2 but with less privacy).

Creativity Might Not Be Enough

I'm only scratching the surface of Web3. However, looking at these small projects, it's easy to see why so many people think the Web3 ecosystem is so great. I don’t think it will get us off centralized platforms, I don’t think it will fundamentally change our relationship with technology, and I think privacy issues are already below the bar for the internet (which is a pretty low bar!), but I also understand why geeks like me are excited to build for it. At the very least, it’s something new on a geeky level: it creates space for creativity/exploration reminiscent of the early internet days. Ironically, part of that creativity may stem from the limitations that make Web3 so unwieldy. I’m hopeful that the creativity and exploration we see will have positive results, but I’m not sure it will be enough to prevent all the same dynamics of the internet from unfolding again.

If we do want to change our relationship with technology, I think we have to do so intentionally. My basic thoughts are roughly as follows: 1. We should accept the premise that people won't run their own servers by designing systems that distribute trust without distributed infrastructure. This means that architectures anticipate and accept the inevitable consequences of relatively centralized client/server relationships, but use cryptography (not infrastructure) to distribute trust. Even though Web3 is built on "crypto", I'm surprised how little cryptography seems to be involved! 2. We should work to reduce the burden of developing software. Currently, software projects require a lot of manpower. Even relatively simple applications require a bunch of people sitting in front of a computer for eight hours a day, day after day, never ending. This wasn’t always the case, and there was a time when 50 people working on a software project wasn’t considered a “small team.” As long as software requires such concerted energy and so much highly specialized human attention, I think it will tend to serve the interests of the people sitting in that room every day, rather than what we might think of as broader goals. I think changing our relationship to technology may require making software easier to create, but throughout my lifetime I’ve seen the opposite happen. Unfortunately, I think distributed systems tend to exacerbate this trend, making things more complex and more difficult rather than simpler and easier.

Original link: https://moxie.org/2022/01/07/web3-first-impressions.html