With the recent Initial Coin Offering (ICO) craze, one cannot stop thinking if all those tokens are really required for the applications or if it’s just a way to raise capital (crowdfund).
While I appreaciate new ways of funding Internet protocols and decentralized applications, there are drawback of doing it this way, one of the main drawbacks is increased transaction costs, especially if the tokens don’t come with their own network. Also, if the software is open-source, what blocks people from taking the code and just removing the token and either getting rid of it completely or just replacing it with the network’s native currency (like Ethereum or Bitcoin, depending on the underlying network).
Most of the tokens are an afterthought to raise some money for a decentralized project. This part is nice and I wish the authors of these projects get paid something.
There are a few drawbacks to this. First of all, the users of these new projects still need the network’s native currency (Ethereum, Bitcoin, …) to pay for the transaction costs. So the user of a decentralized application needs to buy some token, they usually don’t know how much in advance, so they will probably overpay. Then they need a wallet with the native token to make any transactions on the network.
Let’s imagine we have a decentralized ride-sharing app and a “kilometer” token to pay for the kilometer (it is divisible into 1000 pieces, so you can pay for each meter of the ride). In order to call a car on-chain, both parties also need to have Bitcoin (if they run on Rootstock) or Ethers (if they run on Ethereum). So we have a user that needs to buy two weird currencies on the market. And there is currently no way to remove the native token (you can buy the “kilometer” token on a decentralized exchange on-chain, but you cannot do it the other way).
So the first problem is transaction costs. The second problem is the size of the network. Since all users need the native token as well as the “app-specific” token, the network of native token users is by definition larger. So in effect we are reducing the network effect – just because we wanted to do a crowdfunding in the first place.
The token also adds complexity to the smart contract, which increases an attack surface. If there is no token, people can safely store their Ether and contract itself not owning any value is a good thing. If it goes wrong, we just upgrade the contract (trustlessly), release new version of application, copy the state (like reputation of the drivers and users) and go on. You can partially do this with the tokens, if you can identify the problem.
Advantages of tokens
Of course, there are also advantages. First of all, the tokens are a way to finance the initial development. I will mention other ways of doing that.
The other advantage is that the token allows for dynamic pricing of the app’s value (you can trade the “kilometer” token on the market and it will automatically adjust according to supply and demand – although in this case, the adjustment is not such a good idea, because in this case the supply and demand is based on current traffic, there’s a difference between the value of one “kilometer” at 8am in the traffic jam and at 4am).
Another advantage is an indication of demand for the service. I would say that this advantage is not as good as it might seem, because it could be just a pure speculation on all cryptos and ICOs, which is what we’re seeing now. There is much less demand for people to actually implement the apps than there is a demand for buying a token and selling it later. It’s not a bad thing in itself, but I would take the information about demand coming from a token offering with a grain of salt. A big grain of salt.
Detokenizing means taking the project’s open source code apart and removing the token. If there’s incentive mechanism needed, we cold replace that with the network’s native token (bitcoin, ether) which is required anyway to pay for transaction fees.
Half of the really useful projects I saw don’t need a token and it actually makes things more complicated and does not add too much value for the users.
Detokenizing means reducing the transaction costs.
First of all you could argue how the projects are going to get funded? Well, if the code is open-source, it’s not my problem. Here’s an analogy: Red Hat was creating Red Hat Linux. Someone (called a Centos project) took the open-source code, removed all the trademarks and recompiled and released it for free. They didn’t ask the Red Hat’s investors if they could do that and this possibility came from the fact that the code is open-source. (BTW: Centos has been acquired by Red Hat later and they still continue with this strategy of releasing “free” versions of compiled Red Hat code).
So by making a decision to open-source all your code (and whitepaper, documentation, etc.), you are allowing people to do this – explicitly. You have an option of not releasing the code (at least for the client app), but that is usually not a good decision in the smart contract world, because people don’t tend to trust some random code with their cryptos.
By detokenizing, most projects would gain security, they would get better network effect (removing transaction costs in entering the network).
Update December 26th 2018: This happened exactly as I predicted. I wrote this article with ZRX project in mind. While writing this, I was actually looking at ZRX code and trying to see how hard it was to remove the token while keeping the protocol still functional. I found out that it is possible, but it’s too soon to do it, because the ZRX protocol was not mature enough yet. Now, DDEX is forking 0x to remove the ZRX token.
Other ways of funding the project
So how are the projects supposed to get funded if someone is detokenizing them? What is the alternative?
Crowdsale without token
You can still do a crowdsale. With the decentralized ride-sharing app, you can sell ride discount credits and have a simple token outside your main smart contract that handles the tokens. You can then allow the discount contract to apply a discount per transaction. People would still pay with the token’s native currency, they would only get a discount for the credit. After some time (the credits could even expire), you would be fully converted to the native token of the network and you can kill the discount contract. Or do another crowdsale for new features.
How is it different? Well, first of all, it makes much less sense for someone to fork your code, they would not gain much. Also, you are keeping the native currency’s network, because new users that did not participate in crowdsale can just pay with the cryptos they are already using. Lower transaction costs – that’s what this new thing is about anyway, right?
Also, you could limit the transferrability of the tokens, making it difficult to create a secondary market. This way you get information about the demand for the service, because speculation would be much more difficult. You cannot easily transfer the token to third parties. You can trade your private key, but that’s all. You can of course allow the trading, if that is what you think is better for the project.
Also, you can just charge fees for your service, as any other company out there.
Token appreciation bet
If you are going to build a killer app that everyone wants to use, you can get a venture capitalist (fund) to finance your project. Or you can do this bet yourself. Here’s how this works:
You use half the money you get from the VC to pay for the staff and to develop the project itself. You use the other half to buy the native token.
When you are successful, you provide a killer app that will cause a massive rise of value of the native token (that’s the bet).
To make it clear, let’s say I create a killer decentralized app that everyone wants to use. This would cause appreciation of the token value, because people need it in order to pay for transaction fees and any other cost of the application.
This model has some interesting properties, notably – if someone improves on the project and creates a better version based on your work, they are not your competition. The vcs still get a return of their investment, even if the return comes from a competitor.
So what would normally be your competition becomes your colleagues. You are decentralizing innovation in a way that makes you embrace competitive experiments. This supercharges the market forces.
So you can invest in the native tokens, still create great projects, add value to the ecosystem and get a nice profit out of it. You can support the whole ecosystem, help other companies get their apps done too, because they too will grow your investment.
Why would the VCs invest in this way? Why not just buy the token? If your project makes real sense and adds huge value to the whole ecosystem, you will create appreciation that would not happen otherwise. And you still get a hedge against someone just stealing your idea (or some implementation that did not yet reach a product/market fit) and doing it better. These two cases would still not happen if the company did not exist.
Make your competitors work for you, make profit from the value added by other projects and build this crypto utopia together.
This model of course works only for huge projects that add a lot of value. But there are a lot of these that still need to be done. We are building a parallel, decentralized infrastructure. That includes a parallel financial system, parallel shared economy, prediction markets, notary services, etc. If you think this is the future and you can help building it, buy some cryptos and go for it. If you need to fund the development, make it work (VC money, investors, …). If this is the future, which many of us believe it is, it does not matter if any one of us is successful.
There’s a lot of work to be done. Less ICOs, less tokens, more UX, less friction, lower transaction costs, better support for smartphones(!) and a brighter, decentralized and free future awaits!