The current major cryptocurrency is Bitcoin. Changes to the Bitcoin protocol have to be conservative and slow so as not to destabilize the network, and thereby diminish its value. This in turn prevents rapid innovation that is prevalent in software development. The original solution to innovate upon Nakamoto’s invention was simply to fork the Bitcoin software, tweak it as desired and release a new system. Namecoin was the first such fork. It introduced features that were not present in Bitcoin, such as an alias system, a .bit top level domain registration and DNS service. Not too long after that developers started releasing Bitcoin-based clones on a regular basis. In late 2013, early 2014 there were upwards of a 100 such clones introduced daily. Most of them modified only a few parameters, (e.g. block time, hashing algorithm, number of total coins, reward schedule, etc.) and did not offer any significant improvements.
While such experimentation should be encouraged, there are a number of problems that arose as a result of the proliferation of these alternative blockchains (altchains). Each of the altchains had its own native token (coin), in order to experiment with the altchain a user had to either purchase the tokens, or acquire the appropriate mining hardware. The ability to sell newly created coins on these altchains attracted dubious characters trying to make a quick profit. The term scamcoin was soon formed. Security is another problem that plagued some altchains, including low network hashrate allowing an easy 51% attack. Clearly some kind of solution was required. Innovation must be allowed and encouraged, but not at a cost so great that it might damage or even destroy the cryptocurrency industry as a whole. To that end a group of researchers formed the concept of pegged sidechains.
As the name suggest sidechains would exist alongside some main chain. Again, currently the main chain would be Bitcoin (but that could change in the future). At its core pegged sidechains would be a technology that allows movement of assets (coins, tokens, etc) from one blockchain to another. The authors of the sidechains white paper identify five properties that are desired for pegged sidechains.
- Assets can be returned to the ‘parent’ chain only by their current holder.
- There should be no ability for a malicious party to prevent a transfer from occurring.
- Transfers should be atomic (completed in its entirety, or not at all).
- Bugs in one sidechain must not affect any other chains.
- Sidechains should be fully independent, with any external data required supplied by users. Reorganizations should be handled cleanly at all times, and disruption should remain on the affected chain only.
The above properties deal mainly with security issues. But what about the more practical matters? What do pegged sidechains offer over and above the status quo? For one they allow the creation of a sidechain without all the negativity that surrounds brand new altcoins. Unless there is a rather well known team behind a new project, generally it is not looked upon favourably. Rightly so, the altcoin cemetery is full of not so well thought out altcoin corpses. A sidechain could be seeded by bitcoins with a known value. For example, new coins on a sidechain could only be created by transferring bitcoins at a predetermined exchange rate. Of course it would be still possible for a sidechain to have its own token that would be traded just like an altcoin. Currently existing solution involves sending bitcoin to a publicly verifiable, unspendable address. This amounts to the destruction of some funds, however, it can be thought of as a transfer rather than a destruction. The authors of the sidechains whitepaper call this a one-way peg.
The ability to create sidechains allows for permissionless experimentation with the protocol. For example a side chain could be created with much bigger block size allowing, as a result, more transactions per time frame. Bigger block size is something that Bitcoin could use, but experimentation on the Bitcoin chain may be costly if the change proves undesirable. Similarly, other features can be experimented with on sidechains and moved onto the main chain only when proven to work in practice. It is also conceivable, although not very likely, that a sidechain could overtake a main chain in popularity and therefore itself become a main chain.
The main innovation of sidechains is the proposed two-way peg. Assets can be transferred from a parent chain to a sidechain (for example, from Bitcoin to a newly created sidechain), the original assets will be now locked on the parent chain. Because of security concerns there must be a confirmation period, once it passes, the coins can be moved onto the sidechain where they will be subject to a further delay in availability called a contest period. Both waiting periods are necessary to ensure integrity of funds on both the parent, and the sidechain. As mentioned above such a transfer is currently possible under the proof of burn concept. The new feature of sidechains is the ability to transfer the asset back from a sidechain to a parent chain, thereby completing the two-way portion of the peg. If a sidechain were ever to consider shutting down the assets can be transferred back, at the same exchange rate (assets can be transfered back at anytime). This way a user wanting to try out a new sidechain because of some new features does not stand to lose all of her initial investment.
Altcoins suffer from low network hash rates, making themselves vulnerable to a 51% attack. Merged mining is an existing solution to that problem. In merged mining a pair (or more) of blockchains are mined at the same time. Likewise a sidechain’s creators could take advantage of Bitcoin’s security and utilize its hashing power via merged mining. Another improvement in security for a sidechain would be to also keep track of the parent chain. This simplifies the transfers from the parent to the side chain, but at an additional cost of keeping track of the main chain. This is called an asymmetric two-way peg, but the authors do not explore it in detail.
Sidechains of course are not free of drawbacks. The main one is the added complexity to the network, to keeping track of assets, and to user-end software. Another is the need to deal with and prevent fraudulent transfers. Centralization of mining is also an issue, as more and more resources would be needed to keep track of multiple chains, smaller miners might elect not to participate. Finally, risks of a soft-fork is somewhat increased in a sidechain as they have a different level of security when it comes to transfers. The last risk can be alleviated, but at a risk of losing isolation which was one of the main motivating features behind sidechains.
Pegged sidechains are an interesting technology still in its concept stage. The technology will make it possible to transfer assets from one chain to another and then back again. Hence it is called a two-way peg. The main contribution of sidechains would be the ability to experiment with the Bitcoin protocol in a permissionless way, but with the added legitimacy of assets being backed by already established assets. There are some drawbacks associated with sidechains but they are not unsolvable. We are looking forward to seeing this technology in action.
The author would like to thank the following colleagues for feedback: Carla Miller and Paul Jones.
 S. Nakamoto, Bitcoin: A peer-to-peer electronic cash system, 2009, https://www.bitcoin.org/bitcoin.pdf
 [Bitcoin-development] is there a way to do bitcoin-staging?, 2013, Mailing list post, http://sourceforge.net/p/bitcoin/mailman/message/31519067/