Problems on the Blockchain
Play audio recording
Published: 12 Dec 2015 Tags: Article
Blockchain technology (the best known of which is the cryptographic currency Bitcoin), has inherent positive aspects including decentralisation and anonymity. The technology behind Bitcoin, Ethereum, Omni and others, the Blockchain, in its initial implementation heralds democratic and arguably ecological issues. I'm going to discuss briefly the pitfalls of the blockchain in some of its implementations and go on to review some of the other solutions that have or could be implemented.
Presumptions: at this point I'm presuming that the reader has a basic knowledge of Bitcoin and the concepts in the Blockchain technology.
As part of any Blockchain implementation a ledger is formed to store information concerning transaction history usually with meta data which refers to the blockchain itself along with the specific transaction information (e.g. Jon paid 4 Bitcoins to Phil). The ledger is distributed throughout the network giving the system resilience to tampering. However, large scale cryptocurrencies like Bitcoin can end up receiving over a thousand transactions per block in the chain, increasing the overall size for the ledger for everyone who has a complete ledger copy.
If the size of the network is large and there are lots of transactions occurring the size of ledger can bloat hugely to over 20GB or more for the whole ledger. When distributed a million fold (an arbitrary number of miners) your 20GB ledger now doesn't seem so small at 20 Petabytes.
For a small Blockchain network of tens or hundreds of miners you can still benefit from the resilience of the network without pointless duplication. It is also possible to compress further or trim the ledger in different ways to get the size down. Most crypotocurrencies however, haven't yet moved to smaller chains and ledger sizes, this in combination with the other elephant in the room creates for a harmful cocktail.
Part of most Blockchain technologies a proof of work is required by miners (a computer crunching numbers to verify a new block in the Blockchain). As part of Bitcoin's implementation the difficulty of this proof of work gets more difficult in line with the time it takes to accomplish. This means over time the blocks are added at a more or less steady time interval, even with more and more miners with increasing fast processors competing on each new block.
The incentive to compete in mining for bitcoin is part of what gives it some value as there has been real world effort (in this case energy and time). As outlined in length in a Vice Motherboard article the size off this endeavour is starting to consume more effort than is sustainable, not only for most peoples wallets, but also for the planet generally. Computer farms all of the world have been purpose built to mine for cryptocurrencies producing heat and using more and more electricity to power and cool the servers in their endeavour to create the next block in the chain and be rewarded with coins for doing so, this is a new system of wealth that is anything but ecological.
Bitcoin mining leads to an arms race among miners to grab a slice of the fixed rewards doled out by the network, Stolfi said. The higher the financial rewards, the more miners will invest in powerful equipment to keep up with the competition. Christopher Malmo, Motherboard
The higher the consumption the harder it is for the general public to participate in the spoils, the more energy is consumed to run the mining, which in-effect is an inefficiency of the system which actively asks it's many miners to do the same work at the same time competing against each other. It's a bit like giving all the worker bees in a hive the same job and told to compete to do it instead of working together with no wastage of effort.
The tendency towards centralisation, or as Jaron Larnier would say the creation of a siren server with the proof of work structure is almost against the ethos of the Blockchain itself often heralded as the death of the middle man or the democratisation of money. Ironically, it is already implausible to participate in the creation of Bitcoins and points of entry now seem very traditional: buy a Bitcoin from a vendor with traditional currency or bet on Bitcoin's market fluctuations.
What can be done?
There are already lots of Blockchain technologies with subtle differences that make them more or less efficient inline with the above. As already discussed trimming and compressing can help with storage sizes, the proof or work functionality is a little hard to resolve however. One could manipulate the various variables associated like difficulty, time between blocks, number of nodes working on a block, how much of the network is broadcast to etc. One drastic measure would be to discourage competition altogether giving less incentives to miners allowing for a platform which might be more concerned with the content that flows through it rather than the platform itself.
During my research for this article I came across maybe the best example of Blockchain technology given thought to these issues: PeerCoin. PeerCoin uses proof or stake, where coins have lifetimes that influence their participation on the network, along with proof of work with a limit so that it is easily achieved. Other considerations have also been made for the size of duplication on the blockchain and its size.
Though I don't necessarily endorse PeerCoin it shows that the Blockchain isn't an inherently inefficient technology but there is still thinking and experimentation to be done before we wholesale construct a Blockchain financial system and indeed society. There are Blockchain augmentations like trust algorithms and open source Blockchains that should be explored before we sell out the banks for eco-disaster.
In my personal view without a more efficiant system, Blockchain technologies should be used on smaller networks.