Why blockchain is not such a bad technology
Credit to Author: Alexey Malanov| Date: Thu, 28 Sep 2017 13:00:09 +0000
We recently published an article called Six myths about blockchain and Bitcoin: Debunking the effectiveness of the technology. Now, let’s talk about solving those problems to make blockchain work better than it does now.
Before we start, it is important to remember that blockchain and Bitcoin are not the same thing. Bitcoin technology combines several technologies: money transfer principles, cryptographic principles, blockchain proper, the concept of consensus, the proof-of-work principle, peer-to-peer networking, participant motivation, Merkle trees for organizing transactions, transparency principles, hashing, and more.
Therefore, on the one hand, blockchain problems arising from the form in which it is used by Bitcoin are not universal, and it can work differently for other currencies. On the other hand, right now the market is dominated by Bitcoin-like blockchains based on proof-of-work (POW).
Problem: Blockchain is slow and inefficient
Bitcoin’s throughput is seven transactions per second, not for each participant, but for the whole network. And for Ethereum, the second-best in terms of capitalization, it is 15 simple money transfers and 3–5 smart contracts per second.
The POW principle accepted for most currencies guarantees that electricity consumption and the amount of hardware will grow until mining becomes unprofitable. However, growth of overhead costs never improves the quality of the services provided — it’s always 7 transactions per second, no matter how many miners are there and how much electricity they burn.
The Lightning Network
Experts have long been concerned about the problem of insufficient transaction speed in the Bitcoin system, and to address it, they invented the Lightning Network.
This is how it works — or, how it will work, once it is launched: First, certain network participants who need a faster transaction rate set up a separate channel — consider it a kind of private chat room — and, as a guarantee of integrity, make a deposit in the main Bitcoin network. Then they start exchanging payments separately from the rest of the network — at any speed. When the channel is no longer needed, the participants record the results of the interaction in a public blockchain and, assuming no one violated the rules, receive their deposit back.
Optimistic predictions have the Lightning Network launching as early as this year, enabling millions of transactions per second. So much for “slow.”
Problem: Blockchain is bulky
Blockchain is bulky, but that stopped being a problem after some trust was built on the network. In fact, you don’t have to download and check everything to believe the likelihood of deception is very low.
Web wallets
First of all, existing Web wallets and Web services store everything and do all of the work for you. If no one complains about a certain service, it can very well be considered reliable and somewhat trusted.
It also comes with an important advantage compared with traditional payment systems. If one Web wallet closes, you can simply switch to another one, because they have the same transaction records — blockchain is the only one. Compare that with what would happen if your regular bank encountered a glitch or went bankrupt and you needed to switch banks.
Thin wallets
Satoshi himself described another, more advanced (and more reliable) method back in 2008. Instead of storing and processing the entire 100GB blockchain, you can download and check just the block headers, as well as proof of correct transactions that are directly connected to you.
If many random network nodes that you are talking to report the block headers are exactly the same, you may rather confidently say that everything is correct.
At the moment, the headers of all existing blocks take up only 40MB, which isn’t much. But you can save even more: You don’t have to store the headers of every transaction that ever happened; you could start with a specific moment.
Problem: Blockchain is not scalable
A system’s scalability refers to its ability to improve with the addition of resources. The classic blockchain is indeed completely unscalable; adding resources does not affect the speed of transactions at all.
It’s interesting that the classic blockchain is scalable neither up nor down: If you built a small system for solving local problems based on the same principles, it would be vulnerable to a so-called 51% Attack — anyone with enough computing power could come in, immediately take over, and be able to rewrite history.
Plasma
Joseph Poon (the inventor of the Lightning Network) and Vitalik Buterin (a cofounder of Ethereum) recently proposed a new solution. They call it Plasma.
Plasma is a framework for making a blockchain of blockchains. The concept is similar to that of the Lightning Network, but it was developed for Ethereum. Here is how it works: Someone makes a deposit in the main Ethereum network and starts talking to other clients independently and separately, supervising the execution of his or her smart contract and the general rules of Ethereum on their own. A smart contract is a mini-program for working with money and Web wallets. It is the key feature of Ethereum.
From time to time, the results of these individual communications are recorded in the main network. Also, as with the Lightning Network, all participants oversee the execution of the smart contract and complain if something is not right.
So far, the proposal is just a draft, but if the concept is successfully implemented, the problem of blockchain scalability will be a thing of the past.
Problem: Miners are burning up the planet’s resources
Proof-of-work is the most popular method of reaching a consensus in the cryptocurrencies. A new block is created after lengthy calculations performed solely to prevent rewriting of the financial history. POW network miners burn a lot of electricity, and the number of megawatts wasted is regulated not by safety concerns or common sense, but rather by economics: Capacities expand as long as current cryptocurrency exchange rate keeps mining profitable.
Proof-of-stake
An alternative approach to distributing the right to create blocks is called proof-of-stake (POS). Using this concept, the likelihood of creating a block and thus the right to receive an award (in the form of interest or newly emitted currency) depends not on how much computational work you done (how much electricity you burnt), but on how much currency you have in the system.
If you own a third of all coinage, you have a one-third probability of creating a new block, thanks to a random algorithm. This principle is a good reason for participants to obey the rules, because the more of the currency you have, the more interested you are in a properly functioning network and a stable currency rate.
Proof-of-authority
A more radical method exists as well: letting only trusted participants create blocks. For example, 10 hospitals can use a blockchain to keep track of an epidemiological situation in a city. Each hospital has its own signature key as proof of authority. That makes such a blockchain private: Only hospitals can write to it. At the same time, it helps maintain openness, an important quality of the blockchain.
However, proof-of-authority is detrimental to the original blockchain concept: The network effectively becomes centralized.
Resources can be used for good
Some networks do useful work within the proof-of-work concept. They look for prime numbers of a certain type (Primecoin), calculate protein structures (FoldingCoin), or perform other scientific tasks that require a lot of calculations (GridCoin). The reward for “mining” promotes investing more resources in science.
Problem: Blockchain is decentralized and therefore is not developing
It is not very easy to introduce changes into a decentralized network protocol. The developer can either run mandatory updates for all clients — although that kind of network cannot be considered truly decentralized — or persuade all participants to accept the changes. If a significant proportion of them vote against the changes, the community may split: The blockchain will split into two alternative blockchains, and there will be two currencies. That split is called a fork.
Part of the problem is that different participants have different interests. Miners are interested in growing rewards and interest; users want to pay less for transfers; fans want the cryptocurrency to become more popular; and geeks want useful innovations to be added to the technologies.
Two of the largest cryptocurrencies have already split. It happened with Bitcoin not too long ago, when participants were unable to agree on a strategy for expanding block size. A little earlier, something similar happened with Ethereum, the result of a disagreement about if it was fair to cancel a crack on an investment fund and return the money to investors.
How can such situations be avoided?
Tezos
It is possible to encode into a cryptocurrency the ability to vote on modifications. That’s precisely what the cryptocurrency Tezos, which is about to go on the market, did. Primary voting characteristics are as follows:
- The more cryptocurrency you hold, the more voting power you have. Mining power is irrelevant.
- A vote may be delegated to someone who understands the subject of the current vote better than you do.
- Developers are entitled to a veto for one year after launch, and if necessary veto power can be extended.
- The initial quorum will be 80%, but that can be changed to conform to actual user activity.
It’s thought this approach will significantly reduce the emotional level and the necessity for hard forks.
When voting on these principles, at some point the majority could well eliminate the minority’s voting rights. In short, the rich may take over. However, Tezos’s developers think that such a takeover would have a negative impact on the value of the currency and therefore is unlikely. We’ll see.
Problem: Blockchain is too transparent
Imagine you’re WikiLeaks and you get donations in bitcoins. Everyone knows your address and how much you have, and when you try to convert your money into dollars in the exchange, then law enforcement will know how much you have in dollars.
You can’t launder your money in Bitcoin. Dividing up the money into 10 wallets only means having 10 accounts associated with you. There are services called mixers or tumblers that move around large sums of money for a fee, to obscure the real owner, but they are inconvenient for a number of reasons.
CoinJoin in Dash
The creators of the cryptocurrency Dash (the former Darkcoin) were the first to try solving the anonymity problem, by using the PrivateSend function. Their approach was simple: They designed a tumbler right into the currency.
There were a few problems. First, if someone (e.g., law enforcement) controls a significant number of the nodes that mix “clean” money with “dirty,” they can observe the transfer. Perhaps an unlikely scenario, but still quite possible.
Second, mixing dirty money with clean makes all of that money look a bit dirty — or “gray.” But for gray money to appear clean, all participants have to use mixing all the time.
CryptoNote in Monero
A more reliable approach was invented: a truly anonymous currency called Monero.
First, Monero uses electronic signatures that permit a group participant (designated by the cell) to sign a message on behalf of the group and also prevents anyone from ascertaining who signed it. This ability permits the sender to hide their own traces. At the same time, the protocol prevents double spending.
Second, Monero uses not only a private key for money transfers, but also an additional private key to see what has arrived in your wallet, making it impossible to see someone else’s transaction history.
Third, some senders may want to generate one-time wallets to keep money that is private and funds coming in from the markets separate. (This recommendation was made long ago over at Bitcoin.)
Conclusion
Our short overview of issues that some talented people have turned to their benefit has come to a close. We could’ve written much more about smart contracts at Ethereum, the bright future of Ripple, or cryptocurrencies without blockchain such as IOTA.
Strictly speaking, the title of this article is inaccurate. We discussed blockchain’s add-ons, not blockchain itself. But that’s the beauty of blockchain: It inspires people to look for ways to improve it.