> When Ethereum/ another crypto VM becomes fast enough, decentralised applications will be the new iPhone - stuff we can’t imagine will be built there.

In other words, "we have no idea, trust us".

The use case is "stuff we can't imagine"?

Imagine someone told you in 1989 when all the internet did was email and ftp, that the internet was going to do a lot of other things but they didn't really know what yet.

The difference is that email and FTP on their own were really useful.

There was still plenty that could be imagined in 1989 that the Internet could be capable of - there was none of this "it is just too unimaginable to even consider" nonsense.

E.g. there were plenty of dialup BBS services and things like Prodigy with GUI, chat, etc. in the 80s. It wasn't that much of a great leap to see how moving those types of things from proprietary networks to the Internet would occur.

DAOs are fantastically interesting and if you look into the ones that work, Vitalik’s ideas for Quadratic Voting, there’s a lot toucan imagine could happen there.

DAOs and NFTs are already incredible in my view, they are just in pre-alpha stage and everything is kind of a mess. But people see the pre-alpha and goes “this will never work”. I really disagree. I see for instance how DAOs and blockchain could help developers receive compensation for commits to open source software, creating a possibility for the commercialisation of open source that competes with even large companies. I know people who looked into this. And I hope it happens.

It would dramatically change the incentives if you could get as much or even more money from open source software as a developer, and even more so if you could colect royalties for your work wherever it’s used. The entire digital economy would change if you could have open source software that’s fully open, permission less and paid for.

> I see for instance how DAOs and blockchain could help developers receive compensation for commits to open source software, creating a possibility for the commercialisation of open source that competes with even large companies. I know people who looked into this. And I hope it happens.

I was appropriately reprimanded by dang yesterday for starting a comment with an attack, so I'll slow down and just point out the following:

1. GitHub, for example, already makes it extremely easy to contribute to open source developers. It's called GitHub Sponsors.

2. I have actually donated to a developer that writes a library I love using GitHub Sponsors. It was trivial to do, I just entered my credit card, the money came out on a monthly basis, and I cancelled after a year (I told the dev I'd support them for a year). I'll also note that I don't live in the same country as the developer and the developer's country uses a different currency.

3. The thing that gets me about these kinds of examples proposed by crypto fans is that the mechanics of how to, for example, pay open source devs is zero part of the problem. The problem is that a lot of people are willing to work for free, and software users don't like to pay for stuff that they can get for free.

Again, I see a lot of "hand-wavy utopianism" from crypto enthusiasts, but I see zero examples of (a) why this would be better than GitHub Sponsors, or (b) why do you think GitHub Sponsors hasn't been more successful (hint, it's not that actually making payment with a GH Sponsorship is difficult).

I’ve gotten some flak from dang as well ;) happens sometimes

I think we are describing fundamentally different economies. GitHub Sponsors is optional. I’m talking about a way to have mandatory payments in open-source, with fully transparent automatic payment mechanisms to all contributors.

And before you say well how, MakerDAO is kind of already working like this (I’d have to look into it more but that’s my impression).

Imagine you buy a piece of software - and it costs you 20$. It’s open source but when you buy it you get an NFT that lets you run the software. The commit tree and their importance are valued by the community. And when you buy the software everyone who’s contributed gets a piece. If you add or extend to it, you’ll also get a share, or you can sell your modules on the side and make your own money on top. But using the blockchain as a method of developer verification, open source becomes profitable in a way that’s verifiable, decentralised and permission less. Anyone can fork but whatever much of the software youve re used still goes to the original developers.

I’m talking about full open source software that would always be paid, where the devs would be IDed on chain, where anyone could write new modules for and get paid on chain for their contributions. All of this is not crypto Utopianism - again some DAOs are already there or close.

Im not talking about “hey sponsor this dev”. I’m talking about a way to earn money while keeping software open source and rewarding people for expanding it in a completely novel way. This is already partially true, even if my whole vision at the moment is a bit fuzzy, it would solve a major challenge that I think we have now.

I am a bit sleepy so I wonder if my comment will make sense but I hope it does.

But fundamentally I believe MakerDAO already does this, it’s just a web app and you pay to use it. Part of that revenue goes to Ethereum miners/stakers, some goes to the protocol, and the protocol decides who to reward.

I don’t see why that model wouldn’t work for a lot of other things. Essentially with crypto VMs you can enforce a model that A. Is open source B. Rewards contributors financially.

Asking why this is better than GitHub Sponsors is a bit like asking why people didn’t pay for digital art before NFTs. It’s just fundamentally different imo.

There is a lot going on here, but I'll just say I don't see any benefit in reality:

1. "It’s open source but when you buy it you get an NFT that lets you run the software." You seem to fundamentally misunderstand what open source software is. If you need to buy some sort of NFT to run your software, it's not open source, and this goes against pretty much every definition of open source I've ever seen. You're free to sell your own software now, just don't call it open source if it can't be modified and run independently.

2. "But using the blockchain as a method of developer verification, open source becomes profitable in a way that’s verifiable, decentralised and permission less." This is another pet peeve of mine, when crypto enthusiasts rediscover cryptographic principles that have already been in use for decades and decide by calling it "blockchain" that it is something novel. We have a very nice, cryptographically secure method to verify the commit history of any open source software. It's called git and it's been in use for about 2 decades now.

Again, all I really see is a rube-goldberg machine trying to solve a problem that is fundamentally not about difficulty of existing payment systems.

I agree with you. Of course you could have a GitHub Sponsors that would work fundamentally similar to a DAO. Then the question is - why do you think it doesn’t?

Data immutability and decentralisation helps this use case from my point of view for a very simple reason - because if you’re a developer it means no one can in the future just remove you from the list of people who will get money for contributing code in the past, and that users/community can determine to approve your contributions to the repo instead of some centralise core team, who may or may not decide to let you reap the rewards of the repo.

If your code is being used, you get paid period. The community gets to vote on who gets paid for what, certainly, and you can easily prioritise features in the roadmap by just letting people place their tokens/rewards on certain features and let developers figure it out.

Maybe the NFT idea is pretty silly. Actually what I should have said is - MakerDAO already built this and I can see it being applied to other pieces of software. The system works, so far. Users get tokens, they use those tokens to reward developers for their work, they can ask developers to prioritise a feature by pooling their tokens as a reward, any developer, core team or not, who solves that feature gets paid, and any significant contribution as determined by the community gets to reap protocol rewards for a while.

Look at the synergy you can have between the users and developers, and look how it’s all being done in a fully decentralised and permission less way. Anyone can contribute, users pay for features, and everyone reaps protocol rewards for as long as its used. Could GitHub sponsors do that? Maybe. But a DAO might actually be a better way just because it’s trust less.

Now I agree with you it’s Rube Goldberg-y. The thing is blockchain solves trust issues, not tech issues. A DAO with reasonable rules sounds inefficient, but it’s more capable and adaptable than any centralised solution.

I’m not sure why it’s so hard to understand. Basically you are building an open source piece of software where people get paid to contribute, irrelevant of their background, location or previous history. Users get to be in control and developers get to be paid to make open source software. MakerDAO already does this. Again - is this something Sponsors can do? And if it is, then why doesn’t it?

Isn’t the idea that any developer can make a commit to any piece of open source software and get paid for it, not only once but recurrently, far more interesting than our current system of “work for company get paid”? Isn’t it phenomenally more interesting if one could live their life looking around for interesting things to do, fix them, and get rewarded for it?

Let’s imagine two futures and an app that’s just been bought - Figma. Figma is a cool app that people worked on, did a good job and the owners (repo owners essentially) sold to Adobe. The whole software is closed source, the developers got paid per time worked, and only 2/3 people made the decision to sell the repo to Adobe. Lots of devs there might not want to work for Adobe, but did they have a choice? The repo has been acquired and them with it.

Now let’s imagine FigmaDAO. Same piece of software but it’s open source and run by a DAO. One day you get really interested in image manipulation and rendering. You look into the FigmaDAO feature request/reward roadmap and you choose to implement a much needed feature. The community votes to include your work, and you get paid the reward. They also decide that your feature addition is so important that it will earn you 1% of protocol revenue for 10 years. You are now part time benefiting from FigmaDAO. Suddenly you realise there’s another DAO, GimpDAO where you could use your newly acquired skills to build another much needed feature. You move into that community, deliver the feature and… rinse repeat.

On one of these futures, we carry one as we are and companies keep building piles of closed source software. On the other one, developers become free agents who get to contribute to any piece of software at will and get rewarded for it, without having to go through bothersome tech interviews and just building (permissionless), their contributions are recorded in an immutable ledger that no one person in particular can edit (trustless) and they get paid automatically whenever their contributions are used (programmable money).

Which one of these futures sounds more exciting ?

> all the internet did was email and ftp

In retrospect, that would have been a nice stopping point...

> decentralised applications will be the new iPhone.

What makes you think that? Genuinely curious. I can think of lots of reasons that the opposite might be true, so I’m curious as to what my blind spots may be.

It’s a long discussion.

Unfortunately in order to learn about crypto you have to ignore virtually everything that exists in the space right now, since white collar criminals have realised this is a free for all space where they can commit fraud at will and never get charged (with some exceptions).

Really there is only one interesting thing you need to study. And that’s Ethereum. Bitcoin is pretty much stuck in the past and it’s a mummified open source project imo. And all the other interesting chains are just Ethereum clones with some changes - Polkadot, Algorand, Tezos are perhaps the very few that matter. But everything comes from Ethereum, and if you study and earn about ETH you will understand the whole space.

So look at the interesting ideas and read the books from the good people who’ve done their research - Cryptopians came out recently and that’s a good one even though I haven’t read it I trust Laura Shin (her podcast is also excellent ). Out of the Ether is also good.

But here are some thoughts:

- DAOs are already working in making fully decentralized open source dev collectives and rewarding them for work done. See MakerDAO

- dapps running in cryptoVMs can be made to run at no cost to the developer. That’s pretty cool, since a lot of the “moat” that Uber, FB and Twitter build is two-fold - first they own your data, then they build giant infrastructure to serve you the app. With crypto VMs you don’t need this - people pay for usage. If you could figure out an algorithm that’s better than Uber’s at allocating people who want to drive point to point, you don’t need to go “oh well I guess I go fund raising to build the infra”. Crypto VMs give you that for free, since the user pays for the usage of the application in the transaction.

- decentralised applications can be made safely, even though again they are hard to make and lots of them are just copy paste by fraudsters and schemers (and inevitably get hacked). Uniswap, Compound, Lido, Curve Finance. There are multiple dapps that have never been hacked as far as I know.

- Gitcoin is already quite an incredible little way to fund projects, I believe it’s implemented some form of Quadratic Voting, which Vitalik Buterin has talked about

- NFTs actually work - in the sense that there are real artists building communities and selling their digital art (not monkey pictures) I know of multiple artists who quit mind numbing work at big Corp and became quite successful in NFTs. Again 99% of the space is money laundering, fraud, wash trading - but so is real art (the idea that NFTs are special in that regard is a fiction only someone who hasn’t interacted with the “art world” can hold).

I can go on but illl have to think. But DAOs, NFTs, decentralised VMs with immutable data are pretty cool already. The reason I say I can’t predict what’s going to happen is because even being a digital artist myself, I had no idea about NFTs. And when I discovered them I was blown away. The possibilities that NFTs open up as a digital artist are pretty incredible - you could in theory make “designer levels” for Mario Maker and sell only 7 copies of the map! Making it a collector’s item. And you would colect royalties forever on chain if people sold your map to each other. That’s pretty cool and just couldn’t be done before.

So I have no faith in my ability to predict the future, but I like the internet better with DAOs and NFTs, instead of likes and digital feudal lords who own all your data.

https://overcast.fm/+YVsUUTgOg - the Blockchain Socialist’s interview with Buterin is really interesting, giving you a nice overview of Vitalik’s thinking about the space atm.

Can you say what are the current bottlenecks that are limiting the Ethereum VM?

How far off into the future is decentralized apps being as revolutionary as the iPhone?

I have no idea. Potentially decades. It could also be a flop, but I made another comment with the example of how DAOs by themselves could change open source development entirely. That's probably the best articulation of what I think so far.

For now that’s true. I’m not giving judgment on crypto VMs for 10-20-30 years.

Right now 90-99% of the space is scams and nonsense, which happened because of the lack of a regulatory environment. Any smart white collar criminal and fraudster with brains knows they can make millions to billions on crypto and never be charged of a crime.

Crypto is in its ARPANET whatever days. It’s purpose is not fully clear, there are a few usable applications (Uniswap, NFTs, MakerDAO) but really it’s purpose will only be realised when the main snags are solved:

- it needs massive throughput - it needs some form of hack resistance, even if that means the ability to reverse transactions - it needs some form of regulations to avoid rampant crime and fraud - UX/UI need to be dramatically improved (I’ve heard good things about Rainbow and Argent Wallets).

Then user facing applications will show up naturally. You just need it to be able to process some X amount of quantity of transactions, and potentially, reversibility.

as opposed to the new penny stock which is what it is now?

The penny stock phenomenon comes from the Bitcoin fantasy of it becoming a currency. Cryptos are not currencies and I don’t think they will ever be. Ethereum is just liquidity for a decentralized VM, and whether that idea will work out or not is decades away from being clear.

Basically it’s programmable money.