Posts

Yes, but can it run Doom? Layer 2 solutions and blockchain scalability.

, , ,

Just like any new technology, especially one as transformative as blockchain, the hype and expectations start to run up against the limits of its capability and developmental roadmap. In the case of blockchain, throw in the birthing of digital Tamagotchi known as “crypto kitties” last November, and suddenly we see the boundaries of those capabilities tested, even before the vast majority of applications go live.

Ethereum’s “Ether” is the second largest in market cap at the time of writing at USD 84 billion. It is also the protocol most widely referred to when mentioning smart contracts or programs capable of executing code on a blockchain. One of the biggest successes in crypto and blockchain concerning its ability to proxy for trust and generate an infinite ledger of transactions has also become its most prominent frustration. Blockchain and cryptocurrencies are, however, based on ideals that transcend the technology. They broadly cover libertarianism and anti-elitism, and in doing so, the ambition to compete with or replace the likes of Visa isn’t the core priority for those developing the protocols. Developers are content to get it right and protect the systems they are working on, with releases, changes and new directions often taking years and through consensus-based processes involving lengthy testing periods rather than shipping quickly. Something entirely dissimilar to the world of corporate technology and it’s broad adoption of agile principles.

There have been lots of discussion of on-chain solutions to efficiency problems, some of which have resulted in actual implementations, others are still in the works. Things like Proof of Stake and Sharding are examples. What is working against on-chain efficiency solutions, however, is the fact that replication to a network of thousands of computers and systems is, by its nature, inefficient. While some of the proposed solutions address this by reducing restrictions on how much data pushed to every node, eventually, the more reductions made, the more potential impact to the security model and decentralised nature of the ecosystem.

Enter Layer 2 Solutions

A question asked is does every microtransaction need to be instantaneously appended to a blockchain or could these transactions spawn secondary chains or protocols with their own rules and governance that are then consolidated and brought onto a central chain like Ethereum. Such concepts are being referred to as second layer solutions and include ideas such as Truebit, Plasma and State Channels. Their primary objective is to help alleviate the burden of having to run data-intensive applications directly on a blockchain simultaneously.

Using the crypto kitties example, if this application developed its own sidechain which oversaw the logic behind when payments and cat creation took place and the necessary mechanism to resolve disputes (or bailing out to the last known state on the central chain), this could operate alongside something like Ethereum and handle multiple processes before committing back once to the ethereum blockchain. Remembering that things like updates that can occur in relational databases are by its nature not possible in a blockchain ledger. But by using layer 2 and sidechain solutions, a series of append commands could instead be condensed into a single Ethereum entry.

One of the downsides in enabling these sorts of solutions is the potential impact on ecosystem stability. The governance and integration required introduces complexity and opportunities for issues to occur. It also decreases the amount of decentralisation by allowing permissioned and private chains to exist which may be centrally controlled.

Travel-related Implementations

Winding Tree are adopting state channels to provide what they see as a temporary solution to a temporary problem. The expectation here is that on-chain scalability progress will eventually catch up, rendering the need for off-chain processing obsolete. This does not include storage however, with the favoured solutions for decentralisation storage data being IPFS (and Filecoin) and Swarm.

Travel Ledger are looking to develop a payment and settlement solution built on the Ethereum network. They have identified an opportunity to deploy blockchain to simplify the complexity of retailer-consolidator-supplier, booking, invoicing, payment and reconciliation processes as the first step in the practical deployment of blockchain technology. Travel Ledger is investigating private blockchain solutions to assist in overcoming scalability issues, but it would be naive not to think that at least in the interim off-chain solutions will be a part of the Travel Ledger platform.

Back to our previous assertion, the winners in this space are the ones who will be able to demonstrate working solutions without holes or the need to wait for infrastructure to catch up. The world of decentralisation is new; however, expectations regarding speed and price have already been set by traditional companies like Paypal, Expedia and Visa. Like Winding Tree, we view layer 2 solutions as an enabler of progress even if only temporary.

Looking forward to the potential of what a shared dataset could mean, the implications are immense. Where, in the past, the customer experience of booking, checking in, passport control and immigration, boarding, customer loyalty etc. would have meant separate systems were needed to integrate and provide each other with information, with a shared, public ledger, all of these systems become interoperable. Along with greater user experiences comes the opportunity to finally create genuinely innovative travel applications.