Sidechains don’t scale — A critique of Bitcoin maximalism

Artist rendition of how Bitcoin maximalists view themselves

tldr; The best arguments for Bitcoin maximalism are that Bitcoin could incorporate desirable altcoin features or feature them as side chains (without native tokens) and so there is purportedly no reason for altcoin to have native tokens. However, Bitcoin prioritizes security and decentralization and is unlikely to become a basket of altcoin features for practical reasons. Bringing the technology of altcoins of Bitcoin would require side chains for security purposes. However, altcoins + atomic swaps provide an even more trust-minimized ecosystem than Bitcoin + side chains (without native tokens) at the expense of a floating conversion rate between the tokens of different chains. Bitcoin maximalists insisting that no altcoins should have native tokens do not understand the security implications or the limitations of this restriction.

What is Bitcoin maximalism?

The opinion, often held very strongly, that Bitcoin is the only cryptocurrency that should exist and that the native token of altcoins are not just technically wrong, but also morally wrong. That is, altcoins unanimously cause harm to the cryptocurrency ecosystem (read: Bitcoin ecosystem) because it is assumed that all cryptocurrencies are competing for the same users, and so every altcoiner represents the symbolic theft of what rightfully should have been a Bitcoiner, and every altcoin transaction should have instead been a Bitcoin transaction. Since the vast majority of altcoins are made for self-serving purposes, out of intellectual laziness no effort is given to discern exceptions to this pattern. Because of this, self-identifying Bitcoin maximalists have given themselves permission criticize cryptocurrencies on the basis of them not being Bitcoin without regard to or understanding of architectural differences, engineering trade offs, and special use-cases addressed inadequately by Bitcoin.

Introduction

The first time I heard of Bitcoin maximalism I thought it was a joke. After learning more about what Bitcoin maximalism really meant, it is fair to say that I wasn’t wrong. Bitcoin maximalism is absurd to the point of being a joke. So is every other kind of absolutist belief where self-identifying in-group participants consider their strongly-held beliefs and aggressiveness towards out-groups to be a sign of their moral righteousness. It’s like they couldn’t get enough divisiveness in modern-day religions and so turned to create their own. They are the social justice warriors of the cryptocurrency ecosystem and every blockchain they see is a prom dress inspired by Bitcoin’s like-ness.

Here I’ll take the more sophisticated explanations for why self-proclaimed Bitcoin maximalists defend this position and discuss their technical drawbacks. I am focusing only on a few topics and there is undoubtedly more to be said for the existence of cryptocurrencies with a native token beyond this article.

The defense of Bitcoin Maximalism

Arguments for this are reflected well by Oleg Andreev’s blog post.

TL;DR: Bitcoin will win the “cryptographic gold” title and every other altcoin imaginable will die. All fancy features like higher capacity, smart contracts etc will be bolted on top of Bitcoin as long as it’s safe to do with all excessive demand satisfied by commercial blockchain networks, separate layers and protocols on the side and on top of Bitcoin.

A refutation of “any feature an altcoin has can always be added to Bitcoin”

Bitcoin has two value propositions: security and decentralization. It is aiming to be the most sound kind of money we’ve ever had. These priorities were the basis of the community schism that birthed Bitcoin Cash. Years prior to the split, the Bitcoin network utilization was so low that blocks were never full, and costs to transact were so low that Bitcoin was considered an alternative for the unbanked in developing countries (people who can be expected to make less than $2 per day). Bitcoin Cash wanted to preserve that low cost to transact, in addition to governance by miners, and supporters felt the tradeoff to decentralization and security was acceptable. Obviously these ideas were not shared by the Bitcoin community. This is not surprising, and in the future that security and decentralization will not be forfeited either.

Bitcoin’s commitment to security is why new features cannot be readily added. Zk-SNARKs, the core feature of ZCash, could never have been added to the Bitcoin main chain being both so new and so complicated. Many other potential features will not be incorporated into Bitcoin because the cost to security is too high, or these features do not have known, provable level of security. A secondary goal of Bitcoin is to be a lightweight network. The Bitcoin main chain was never aiming to be a convenience store of features that could bloat the blockchain or introduce security risks. It is disingenuous to suggest that altcoins have no moral right to exist since there are no technical barriers to to porting over certain features in these open source projects. To incorporate those into the Bitcoin ecosystem would require a side chain which have security considerations I will comment on later.

More importantly, there are many features which cannot be ported to Bitcoin. Architectures cannot be ported. Bitcoin cannot benefit from Nano’s architecture, or Stellar’s, or any DAG. New consensus mechanisms which trade trust-minimization for other properties such as speed or throughput cannot be integrated into the Bitcoin main chain. Again, to be part of the Bitcoin network, they would need to be side chains.

In short, the claim that many features can be added to Bitcoin neglects to consider the security risks and the technical debt / blockchain bloat they are likely to add, so it is likely few to none of them will be incorporated into the main chain even though the features may be useful.

A primer on side-chains

Pegged Side chains are technically a mechanism to transfer cryptoassets between blockchains. When discussed by Bitcoin maximalists, it is often assumed that the side chain will not have a native token. When I refer to side chains throughout the rest of this article, I refer to side chains as blockchains without native tokens and altcoins as blockchains with native tokens. Instead, the idea is for these blockchains to share the bitcoin supply. How this works is a user sends their bitcoins to an address controlled by trusted intermediaries who maintain the side chain. They keep the coins locked in that address and simultaneously generate new bitcoin proxy tokens in a 1:1 ratio on the side chain. Using a similar mechanism, those bitcoin proxy coins are provably burned to release the bitcoins locked by the trusted intermediaries on the Bitcoin main chain.

There are no risks to the Bitcoin main chain since bitcoins cannot be created. The security risk comes down to the ability of the trusted intermediary to censor conversions from the main chain to side chain and vice versa. Even in the event that every side chain and main chain full node disagrees with the behavior of the intermediary, they cannot prevent the intermediary from censoring the conversion of tokens from one chain to another. Furthermore, to ensure that side chain tokens are not being fabricated, any node will need to be monitoring both the side chain and main chain blockchains. I mention this to highlight that there are security risks for the side chain monetary supply (with no native tokens) and higher costs for validation.

The problem with side chains without a native token

The issue with blockchains sharing tokens which may be redeemed in either direction comes down to security. The whole purpose of a native token in a cryptocurrency is to incentivize users to act in ways that secures the system. Bitcoins are distributed as rewards for mining blocks, for example. So what are the implications for a side chain if we disallow native tokens and intend for Bitcoin proxy tokens to be the main (read: only) currency?

1. In the side chain, if there were a block reward for a block producer, the number of side chain tokens would be in excess of locked tokens on the main chain. This would create a fractional-reserve cryptocurrency. Many would find this to be problematic given the long history of fractional reserve banking used in government banking and the recurrence of economic crises caused by it. Any side chain consensus mechanism with a block reward would be severely problematic. This is to say that side chains are poorly suited to experimenting with different consensus mechanisms without breaking the fixed 1:1 conversion rate or forcing a constantly depreciating conversion rate.
2. The security of side chains depends on trusted intermediaries or block producers of the main chain. If you would like to convert your side chain tokens to main chain tokens, you need the explicit permission of the intermediary which has locked up your main chain coins to retrieve them on the main chain. Storing your coins in a side chain is no more secure than keeping your coins with a custodial exchange like Coinbase.
3. If all incentives for side chain block producers came from transaction fees, this would cause a bootstrapping problem. Bitcoin was able to grow in part because of a subsidy called the block reward. This subsidy is designed to support the network in the very beginning to make it profitable to produce blocks before there are enough daily transactions for a fee market to develop. It was expected to take years for Bitcoin’s network effects to grow to the point where it could wean off of the block reward as the primary incentive for the miners to create blocks. Even nine years old, Bitcoin is almost entirely dependent on the block reward to incentivize miners. (A back of the envelope calculation shows that transaction fees are responsible for 3% of the total daily economic incentives. 200k tx per day; $2 avg tx fee; $8k BTC price; 12.5 BTC block reward.) These new side chains are deprived of the ability to bootstrap themselves by incentivizing block producers with a guaranteed reward because it would otherwise create a fractional reserve environment.

Side chains without native tokens can be still useful in limited circumstances

Contrary to my criticism of side chains, they are well suited for certain use cases. If you would like to test out experimental features that could reasonably be added to Bitcoin, a merge-mined side chain could be appropriate. The trusted intermediaries for moving coins between chains could be the last N Bitcoin block producers and this will be the same set of the last N block producers on the side chain assuming wide participation among miners who merge mine the side chain.

Unincentivized validators and trusted block producers could also make for a useful side chain as the trusted validators could be issued by the largest miners among other entities. Ripple, for example, is well suited for this. Having a trusted set of intermedaries for moving coins between chains does not significantly weaken the security of the system since side chain consensus is not any less trust minimized than converting tokens between chains.

More importantly, if side chains do have a native token, then anything can safely be a side chain without threatening the native token monetary supply.

Side chains are often better off with native tokens

The purpose of side chains as described by Bitcoin maximalists is to provide a means to use Bitcoin for circumstances where Bitcoin clearly cannot incorporate certain features on the main chain. I’ve demonstrated that for different consensus mechanisms, side chains without native tokens are an extremely problematic solution because you must either choose between guaranteed proxy token redemption on the main chain (i.e. no fractional reserve banking), or culling block rewards from your consensus mechanism, or breaking the fixed conversion rate between side chain and main chain tokens.

It turns out that it’s actually pretty straightforward to sidestep this problem. With atomic swaps, Bitcoin can be swapped for altcoins in a trust-minimized manner that is much more sound than having a set of trusted intermediaries which side chains require. The obvious downside is that there is not a fixed 1:1 conversion rate between chains but rather a floating conversion rate.

Why self-proclaimed Bitcoin maximalists don’t like altcoins but love the idea of side chains without native tokens

Transfers using altcoins + atomic swaps and side chains are almost indistinguishable with a few key differences I outlined: side chains have a trust-ful token conversion process, side chains disallow experimentation with consensus models, and side chains have a fixed rate conversion between main chain tokens. Beyond these differences, altcoins + atomic swaps and side chains are nearly the same. Given these differences, it seems pretty clear that if you value trust-minimization you probably should support altcoins + atomic swaps.

For someone to support side chains without native tokens in all cases, you have to wonder why the only upside, fixed rate conversion, is so important. Not surprisingly, it comes down to money and Bitcoin’s role in the ecosystem. In the side chain scenario, the only tokens that exist are Bitcoin or Bitcoin proxies. So all the value that gets created through side chains gets captured in the only tokens available: Bitcoins and Bitcoin proxies. In the altcoin + atomic swap scenario, the value of new cryptocurrencies gets captured in the respective token, and this changes the floating conversion rate to Bitcoin.

Conclusion

Perhaps it’s best to recognize that Bitcoin has priorities which will limit features that get incorporated into the main chain, putting aside that not all altcoin features can be appropriated in that way. The idea of altcoins having their own tokens serves important security purposes in many cases and the floating conversion rate from altcoins + atomic swaps are actually more trust-minimized with fewer limitations than what side chains without native tokens can provide. Ultimately, tokens in a properly designed cryptocurrency serve an important security purpose, and it is best not to tinker with that for reasons unrelated to security, such as the price of Bitcoin. Altcoins can still serve as side chains, but to say that bitcoin could take the place of the native token is false when validators are incentived by block rewards.