The DAO, Ethereum, and what bitcoin can learn from them
Posted: Mon May 16, 2016 5:55 pm
While much of the cryptocurrency industry's focus has been on Craig Steven Wright for the past few weeks, The DAO project has been silently picking up steam. With over $100m in DAO tokens sold so far, intelligent observers are paying close attention to this groundbreaking project. In this post, I'll describe The DAO, evaluate its potential for success, and discuss what the unbelievable success of the token sale means for the future of the cryptocurrency industry.
The DAO (short for distributed autonomous organization) is the reference implementation of a concept that has been discussed almost as long as bitcoin has been in existence, long before the Ethereum platform was available to implement it. At its initial, most basic level, it can be thought of as a corporation that is immune to the legal system. People become shareholders in The DAO by purchasing tokens while they are on sale, which is occurring now. After the sale ends near the end of May, token holders can sell their shares to others through exchanges.
As with any other corporation, its owners decide what activities the organization will perform through voting. For The DAO, any owner can submit a Proposal, or a block of code, which is voted upon by tokenholders to determine whether to execute the code or not. Tokenholders can also take their money and leave The DAO to form a new organization if they disapprove of the results of a vote. Curators, one of whom is Ethereum founder Vitalik Buterin, determine to which addresses The DAO's funds are able to be paid.
You may be initially confused, as I was when I first researched The DAO, as to what the organization is supposed to do. Perhaps the most interesting part of the whole project is that it is possible that $150m will be spent before anyone has actually decided what it will be used for. The DAO code does not specify the focus of the project's activities. The project's website, the owner of which is unknown, also doesn't mandate a specific purpose for the organization.
That means that we could find out a year from now that The DAO has purchased real estate and earns profit in rent, or it pays programmers to design websites, or that (as some have suggested) it becomes involved with transportation and electric vehicles. It could get involved in a combination of different businesses. It could spend all its money and go bankrupt. Or, it could end up doing nothing at all.
In fact, doing nothing is the most likely outcome, because of the voting rules governing what Proposals are adopted. Since there will undoubtedly be a huge number of Proposals, almost all of those that don't get axed by the Curators will simply be ignored. "Proposal fatigue" will make it difficult for people to determine which ones are important. To add to that, 20% of the tokenholders must vote before a quorum is achieved. Given that there is no precedent for this type of organization, the actual number of people who will vote is unknown. However, anyone who has ever tried to get people to show up to any sort of event will tell you that posting flyers in a building and hoping that 20% of the people eligible to attend view the flyers and show up is a losing proposition. It's almost impossible to get 20% of people to do anything even if you ask them in person.
And unlike the real world, The DAO is a complex system that requires people to take their own time to do something that is not likely to significantly affect their lives. I believe that Trump is going to become President, and my actually going to vote isn't likely to change that, so my time is better spent elsewhere. Similarly, everyone believes that a Proposal is going to pass or fail without their assistance, and their own vote matters little, so few will vote. How many readers can ever recall responding to one of those annual board meeting proxy requests that come in the mail for a corporation where you own 50 shares of stock? The only difference between the proxy requests and The DAO is that the number of Proposals from the DAO will be far more overwhelming.
To complicate matters, if there actually was a good candidate opposing Trump for President, I can easily understand the ballot and fill in the oval. Anyone who disagrees with the actions of the DAO needs to download the right software and create a specific type of transaction to register their votes. Many of the tokenholders who bought into The DAO craze as an investment don't understand what they own or that they can even vote. Knowledgeable investors may choose to allow an exchange to retain ownership of their tokens for security purposes, and those tokens will therefore abstain.
Finally, there was a theft at the GateCoin exchange earlier this month that resulted in a loss of a huge number of DAO tokens. Private keys associated with other tokens are undoubtedly lost as well. More tokens will become lost over time, but no more will be created. Meanwhile, the quorum requirement can't consider the unavailability of lost tokens. All of these factors combine to make achieving a 20% agreement on any Proposal very unlikely. There is a rule built into The DAO which reduces the quorum requirement to 10% if no Proposals have been passed in a year, but an entire year is too long in the cryptocurrency industry for an organization to remain inactive but relevant. Some other DAO will take its place if that happens, perhaps one that has more lax voting requirements.
Events are demonstrating more and more frequently that "distributed networks" are not the solution for everything. The underlying problem that blockchains try to solve is to remove the ability for bad people to do bad things. Nobody has ever lost money in cryptocurrency because a bad developer decided on a whim to steal it, but leaderless organizations without good people in charge tend to get little accomplished and eventually become obsolete. We've seen how Ethereum is surpassing bitcoin because it has a leadership team that is able to push through changes over dissent when they are for the good of the network. Meanwhile, bitcoin remains stagnant because many bitcoin users trust in "distributed networks" to solve the "good people" problem. I suspect that the end result of The DAO and bitcoin in general, as I indicated in a previous post, is that blockchains are less useful than previously thought, and that it will be proven that a good person needs to be in charge for an organization to succeed.
The lack of an empowered leadership structure is why The DAO, in its initial incarnation, is likely to be beset by gridlock. That said, there is hope - the greatest benefit for the computerized organization is for use by computers.
The DAO's greatest promise is in automated systems that use it to conduct business. An example would be buying and selling computing power on the Microsoft cloud, which accepts Ethereum. With the right Proposal, The DAO can autonomously manage computing power for various uses in order to earn profit. For example, it could see when GPU instances are cheap, and automatically spin up nodes to perform Ethereum mining; releasing them when finished. It could watch for when memory is cheap, and resell nodes for scientific computing that require memory. When the price of GPU power increases, mining could be discontinued and the GPUs rented to other industries. It could automatically manage payments and discontinue customers who fail to pay up. Because its activities aren't limited by humans, The DAO will excel at businesses where APIs are available and which require no human input. These are the only Proposals that have a chance of long-term success.
Bitcoin hasn't caught on for mainstream users because it provides humans little value over human-created money. OpenBazaar won't take off until bots can use it to buy and sell items between themselves. The DAO is likely to be beset by gridlock until humans are removed from the system.
The success of The DAO's launch demonstrates an important truth about cryptocurrency - it isn't built on bitcoin. While bitcoin languishes, the innovation has moved on to other networks. The DAO, or some other DAO if this one's quorum requirement is too high, could become more significant than bitcoin or Ethereum itself. There is no limit to the growth of a DAO, and economies of scale promote consolidation, so a parallel universe probably exists where The DAO will eventually manage the entire world's production.
Whether or not that is likely or desirable, it isn't possible to accomplish in bitcoin. The reason it isn't happening in bitcoin isn't because there is some technical issue or intractable problem of mathematics; it's because there is a variable in the code, some people are opposed to changing it, and no other significant features will ever be adopted until the sides can agree on this one issue. One of the greatest disappointments of the whole blocksize debate is that the best talent, like Gavin Andresen, is wasting its time on Bitcoin Classic. At the very least, I hope that The DAO will convince the best programmers that cryptocurrency is better served by spending their time on development, where their work is respected, and not wasting months trying to convince people well after it has become obvious they will not be convinced. Perhaps The DAO will lead some of them to let the Core developers stagnate bitcoin, and to move on to where the cutting edge now lies.
In conclusion, it isn't an accident that the price of Ethereum is soaring. Part of the price increase is probably the beginning of what I predicted earlier of investors seeking safety for the upcoming bitcoin halving, but The DAO is independently driving huge volume and breaking all sorts of records. This week, even Jeff Garzik began using "bitcoin and ethereum" when talking in the media about examples of blockchains. The DAO itself is probably likely to be beset by gridlock, but it represents something new and innovative that is a technological breakthrough. Even if The DAO is never able to actually agree on a Proposal to do anything, another DAO with more reasonable voting rules will take its place. While the DAO itself is getting all the headlines, the real story is that after a long time when it seemed like nothing was happening, there are huge new applications for cryptocurrency - they just aren't in bitcoin anymore.
The DAO (short for distributed autonomous organization) is the reference implementation of a concept that has been discussed almost as long as bitcoin has been in existence, long before the Ethereum platform was available to implement it. At its initial, most basic level, it can be thought of as a corporation that is immune to the legal system. People become shareholders in The DAO by purchasing tokens while they are on sale, which is occurring now. After the sale ends near the end of May, token holders can sell their shares to others through exchanges.
As with any other corporation, its owners decide what activities the organization will perform through voting. For The DAO, any owner can submit a Proposal, or a block of code, which is voted upon by tokenholders to determine whether to execute the code or not. Tokenholders can also take their money and leave The DAO to form a new organization if they disapprove of the results of a vote. Curators, one of whom is Ethereum founder Vitalik Buterin, determine to which addresses The DAO's funds are able to be paid.
You may be initially confused, as I was when I first researched The DAO, as to what the organization is supposed to do. Perhaps the most interesting part of the whole project is that it is possible that $150m will be spent before anyone has actually decided what it will be used for. The DAO code does not specify the focus of the project's activities. The project's website, the owner of which is unknown, also doesn't mandate a specific purpose for the organization.
That means that we could find out a year from now that The DAO has purchased real estate and earns profit in rent, or it pays programmers to design websites, or that (as some have suggested) it becomes involved with transportation and electric vehicles. It could get involved in a combination of different businesses. It could spend all its money and go bankrupt. Or, it could end up doing nothing at all.
In fact, doing nothing is the most likely outcome, because of the voting rules governing what Proposals are adopted. Since there will undoubtedly be a huge number of Proposals, almost all of those that don't get axed by the Curators will simply be ignored. "Proposal fatigue" will make it difficult for people to determine which ones are important. To add to that, 20% of the tokenholders must vote before a quorum is achieved. Given that there is no precedent for this type of organization, the actual number of people who will vote is unknown. However, anyone who has ever tried to get people to show up to any sort of event will tell you that posting flyers in a building and hoping that 20% of the people eligible to attend view the flyers and show up is a losing proposition. It's almost impossible to get 20% of people to do anything even if you ask them in person.
And unlike the real world, The DAO is a complex system that requires people to take their own time to do something that is not likely to significantly affect their lives. I believe that Trump is going to become President, and my actually going to vote isn't likely to change that, so my time is better spent elsewhere. Similarly, everyone believes that a Proposal is going to pass or fail without their assistance, and their own vote matters little, so few will vote. How many readers can ever recall responding to one of those annual board meeting proxy requests that come in the mail for a corporation where you own 50 shares of stock? The only difference between the proxy requests and The DAO is that the number of Proposals from the DAO will be far more overwhelming.
To complicate matters, if there actually was a good candidate opposing Trump for President, I can easily understand the ballot and fill in the oval. Anyone who disagrees with the actions of the DAO needs to download the right software and create a specific type of transaction to register their votes. Many of the tokenholders who bought into The DAO craze as an investment don't understand what they own or that they can even vote. Knowledgeable investors may choose to allow an exchange to retain ownership of their tokens for security purposes, and those tokens will therefore abstain.
Finally, there was a theft at the GateCoin exchange earlier this month that resulted in a loss of a huge number of DAO tokens. Private keys associated with other tokens are undoubtedly lost as well. More tokens will become lost over time, but no more will be created. Meanwhile, the quorum requirement can't consider the unavailability of lost tokens. All of these factors combine to make achieving a 20% agreement on any Proposal very unlikely. There is a rule built into The DAO which reduces the quorum requirement to 10% if no Proposals have been passed in a year, but an entire year is too long in the cryptocurrency industry for an organization to remain inactive but relevant. Some other DAO will take its place if that happens, perhaps one that has more lax voting requirements.
Events are demonstrating more and more frequently that "distributed networks" are not the solution for everything. The underlying problem that blockchains try to solve is to remove the ability for bad people to do bad things. Nobody has ever lost money in cryptocurrency because a bad developer decided on a whim to steal it, but leaderless organizations without good people in charge tend to get little accomplished and eventually become obsolete. We've seen how Ethereum is surpassing bitcoin because it has a leadership team that is able to push through changes over dissent when they are for the good of the network. Meanwhile, bitcoin remains stagnant because many bitcoin users trust in "distributed networks" to solve the "good people" problem. I suspect that the end result of The DAO and bitcoin in general, as I indicated in a previous post, is that blockchains are less useful than previously thought, and that it will be proven that a good person needs to be in charge for an organization to succeed.
The lack of an empowered leadership structure is why The DAO, in its initial incarnation, is likely to be beset by gridlock. That said, there is hope - the greatest benefit for the computerized organization is for use by computers.
The DAO's greatest promise is in automated systems that use it to conduct business. An example would be buying and selling computing power on the Microsoft cloud, which accepts Ethereum. With the right Proposal, The DAO can autonomously manage computing power for various uses in order to earn profit. For example, it could see when GPU instances are cheap, and automatically spin up nodes to perform Ethereum mining; releasing them when finished. It could watch for when memory is cheap, and resell nodes for scientific computing that require memory. When the price of GPU power increases, mining could be discontinued and the GPUs rented to other industries. It could automatically manage payments and discontinue customers who fail to pay up. Because its activities aren't limited by humans, The DAO will excel at businesses where APIs are available and which require no human input. These are the only Proposals that have a chance of long-term success.
Bitcoin hasn't caught on for mainstream users because it provides humans little value over human-created money. OpenBazaar won't take off until bots can use it to buy and sell items between themselves. The DAO is likely to be beset by gridlock until humans are removed from the system.
The success of The DAO's launch demonstrates an important truth about cryptocurrency - it isn't built on bitcoin. While bitcoin languishes, the innovation has moved on to other networks. The DAO, or some other DAO if this one's quorum requirement is too high, could become more significant than bitcoin or Ethereum itself. There is no limit to the growth of a DAO, and economies of scale promote consolidation, so a parallel universe probably exists where The DAO will eventually manage the entire world's production.
Whether or not that is likely or desirable, it isn't possible to accomplish in bitcoin. The reason it isn't happening in bitcoin isn't because there is some technical issue or intractable problem of mathematics; it's because there is a variable in the code, some people are opposed to changing it, and no other significant features will ever be adopted until the sides can agree on this one issue. One of the greatest disappointments of the whole blocksize debate is that the best talent, like Gavin Andresen, is wasting its time on Bitcoin Classic. At the very least, I hope that The DAO will convince the best programmers that cryptocurrency is better served by spending their time on development, where their work is respected, and not wasting months trying to convince people well after it has become obvious they will not be convinced. Perhaps The DAO will lead some of them to let the Core developers stagnate bitcoin, and to move on to where the cutting edge now lies.
In conclusion, it isn't an accident that the price of Ethereum is soaring. Part of the price increase is probably the beginning of what I predicted earlier of investors seeking safety for the upcoming bitcoin halving, but The DAO is independently driving huge volume and breaking all sorts of records. This week, even Jeff Garzik began using "bitcoin and ethereum" when talking in the media about examples of blockchains. The DAO itself is probably likely to be beset by gridlock, but it represents something new and innovative that is a technological breakthrough. Even if The DAO is never able to actually agree on a Proposal to do anything, another DAO with more reasonable voting rules will take its place. While the DAO itself is getting all the headlines, the real story is that after a long time when it seemed like nothing was happening, there are huge new applications for cryptocurrency - they just aren't in bitcoin anymore.
thinks.