This was originally published on the Let’s Talk Bitcoin blog.
One of the criticisms of Bitcoin I often hear from environmentally conscious people concerns the enormous amount of energy that goes into mining with relatively little to show in return. I have often countered that it is important to not only compare the energy used in Bitcoin mining to the results produced (a very secure network; in fact, one could argue, an oversecured network), but to the system that Bitcoin could potentially displace: the legacy financial system. A redditor did some back-of-the-envelope calculations estimating the energy usage of every bank branch in the US and found that Bitcoin was using just 30% of the energy of U.S. banks while achieving far more efficiency as a payment network and a greater return on investment for “depositors.”
While the efficiency gains are to be applauded, if we are already approaching 1/3 of the energy usage of the U.S. banking system alone at less than 1% of the total value deposited in the worldwide legacy banking system, it would appear that Bitcoin could end up consuming more energy resources than the existing financial system as BTC value grows and mining-related energy consumption grows with it (note: such growth might not be scalable with Bitcoin). This equation doesn’t even factor-in the energy costs of all the altcoins and cryptocurrency 2.0 networks which aren’t merge-mining Bitcoin, or the energy costs needed to produce mining hardware and ship it to customers. Ultimately, it comes down to this: the cryptocurrency community can do better. Those concerned just need to develop and adopt more efficient methods of mining. Luckily, there are a growing number of options for people who want to reduce their crypto-carbon footprint.
For those unfamiliar, mining is the term used for the work done by computers in a block chain-based cryptocurrency network to secure the network against attacks; the possibility of earning cryptocurrency is the incentive and reward for doing this work. In Bitcoin, the “Proof of Work” is generating a cryptographic hash of the current block’s header that is lower or equal to the current target. The target is chosen by the Bitcoin software and changes every 2016 blocks. The reward for producing a successful Proof of Work before any other miners is 25 new bitcoins, and this reward halves every 210,000 blocks. The Proof of Work function appears to lead to a somewhat circular argument: the work is valuable because it secures the Bitcoin network, and the Bitcoin network is valuable vis-a-vis other cryptocurrency networks because it has a large amount of hashing power doing work, increasing the cost of an attack.
(It’s worth mentioning here that there are other reasons for Bitcoin being valuable aside from the large amount of hashing power securing the network.)
Thus far, this explanation has been good enough for most of the Bitcoin community, but others aren’t quite so convinced. Some have imagined combining the concept of mining with “folding at home”- or “SETI at home”-style computational work. Others imagine “proof of excellence,” “proof of bandwidth,” or “proof of burn” as the alternatives for providing security and/ or distributing cryptocurrency units. Determining what is best, or even possible, will be a matter of experimentation. Here are some cryptocurrency projects which are getting creative with the way that their network is secured and currency is distributed:
The first example of a cryptocurrency that implements “useful” mining, Primecoin implements a unique mining algorithm which has miners find long Cunningham chains of prime numbers as Proof of Work. While not useful in the “curing cancer” kind of way, finding long prime numbers is mathematically interesting, interesting enough for rewards to be given out to people who find really long prime numbers. Primecoin miners have thus far set several records for finding the longest Cunningham chains.
Gridcoin is a cryptocurrency that merges scrypt-based Proof-of-Work mining with software known as “BOINC,” or the Berkeley Open Infrastructure Network Computing Grid. Here’s the description of BOINC from the University of California, Berkeley website:
Use the idle time on your computer (Windows, Mac, Linux, or Android) to cure diseases, study global warming, discover pulsars, and do many other types of scientific research.
With Gridcoin, miners can either earn the normal block reward of 5 gridcoins when not running BOINC, or a “subsidy” of up to 150 extra gridcoins per block for maxing out the amount of computing resources dedicated to BOINC.
One of the possible attacks on the BOINC mining method is that miners could somehow trick the program to make itthink they’re solving BOINC problems when in actuality they are not. To head off such a possibility, the Gridcoin developers have issued a challenge to anyone who can trick Gridcoin into giving miners the BOINC subsidy without actually running BOINC. So far, it seems no one has successfully “hacked” Gridcoin in this way.
“Sustainable” Consensus and “Fair” Distribution
Instead of simply adding value to the Proof of Work function above and beyond cryptocurrency distribution and network security, some projects seek to make the consensus process itself more environmentally sustainable, reducing the amount of energy needed for contributing to the network consensus process. This can make for a tradeoff in the distribution scheme: if a project eliminates the need for mining, it eliminates the auto-decentralized method of distributing the network cryptocurrency. This has created arguments about whether or not a cryptocurrency has been or ever can be distributed in a “fair” manner. While it is ultimately the market that determines the value of a network and it’s usage tokens, it is also the values of the network that determine the market for that network. What next-generation cryptocurrency projects are realizing is that eliminating mining and generating coins up-front can create interesting opportunities for incentivizing behavior valued by the network.
Ripple is an open payment network that includes a built-in decentralized exchange and Turing-complete scripting language. Instead of mining, Ripple uses a method called “consensus” to determine the current state of the Ripple network and secure the network against attack. Consensus occurs approximately every five seconds, rendering all transactions which have taken place in that time period irreversible and forever publicly stored on the Ripple ledger. With consensus, the network can be secured by validators using only the normal electricity costs of running a server. Due to this consensus method which uses a ledger chain instead of a block chain, there is no mining in Ripple. Instead, all 100 billion XRP that will ever exist were created in the “Genesis Ledger” and given to the inventors of the network, Ripple Labs, with 20 billion XRP going to the founders of the company (XRP is the native cryptocurrency on the Ripple network and is used as both an anti-spam mechanism and a bridge currency).
In contrast with Bitcoin’s seemingly-egalitarian “one CPU one vote” philosophy, this has been perceived as too centralized by many in the cryptocurrency community, so much so that some have erroneously called Ripple a “scam.” However, Ripple Labs thus far appears up to the task of responsibly distributing their XRP holdings; after all, if this is not done in a way that is beneficial to the Ripple network, then XRP will not gain value and their holdings will be worth far less than they potentially could be. To this end, Ripple Labs has been selling XRP to pioneering businesses and giving it away to early adopters, developers, and charities. The many ways Ripple Labs can distribute XRP is limited only by the creativity of the Ripple community.
Ripple Labs’ latest effort is called Computing for Good and is similar to Gridcoin in that users run software which utilizes spare CPU cycles for solving scientific problems. Problems being solved by Computing for Good participants include analyzing aspects of the human genome, HIV, dengue fever, muscular dystrophy, cancer, influenza, rice crop yields, clean energy, and more. Running the World Community Grid software while signed up for the Ripple Labs WCG team earns a user WCG points which Ripple Labs will redeem for XRP. So far, over 80 million XRP have been given away through the Computing for Good program.
Another potential method for XRP distribution is offering users a “negative interchange” reward. The idea is that payment networks typically charge an interchange fee which adds a cost that consumers must ultimately bear. With a Ripple Labs-funded “negative interchange,” Ripple users would actually get an XRP reward in exchange for spending money with merchants on the Ripple network. This would promote usage of the Ripple payment network and distribute XRP to those who are engaging in economic activity on the network. The developers of the eSpend platform have hinted at such a model for their XRP rewards program; we will have to wait and see whether or not – and to what extent – this method of XRP distribution is actually employed.
Solarcoin is a project started by the Solarcoin Foundation with the goal of incentivizing the use of solar energy. The solarcoin cryptocurrency is 99% pre-mined and the remaining 1% can be mined using scrypt mining equipment (one could mine with a common graphics card found in every personal computer). The Solarcoin Foundation is giving away 1 solarcoin (SLR) for every 1MW of electricity generated by someone who would like to receive “free” solarcoins. All that is needed is to prove to the Foundation that the solar energy was generated using a copy of a utility bill or data from the API of an energy meter hooked up to solar generators (methodology standards are still being developed by the community). The last solarcoins are projected to be mined in the next 20-25 years, reducing the amount of time computers will be mining the coin compared to Bitcoin, which is projected to be mined-out around 2100.
Proof of Stake is a method of cryptocurrency security and generation which uses a miner’s “stake” in the cryptocurrency to determine their odds of finding a new block; put more simply, the more coins you have set aside for Proof of Stake mining, the more coins you earn. It’s like an interest-bearing savings account which rewards you for locking up wealth and helping to secure a cryptocurrency network against fraud.
Peercoin is the first cryptocurrency to implement Proof of Stake, albeit using a hybrid approach where coins are first mined using Proof of Work, with a transition to Proof of Stake over time. The stated intention is to improve sustainability by reducing the energy resources necessary to secure the network. However, at the time of this writing there remain challenges to this method of mining which the developers have promised to address in future software upgrades.
The first cryptocurrency to use a 100% Proof of Stake model, at the launch of the NXT project 100% of the cryptocurrency was distributed proportionally to financial backers’ initial investment. In NXT, blocks are created through a process known as “transparent forging” whereby every node in the network can determine which node will mine the next block, allowing for “VISA/ Mastercard rates of transactions.” Forging new blocks spreads transaction fees among all of the participants in the network who have set aside NXT for mining.
Ethereum is a cryptocurrency project which intends to implement a fully Turing-complete scripting language, making the project an open-source tool kit for next-generation block chain-based services. The Ethereum developers are considering implementing hybrid Proof of Work/ Proof of Stake mining with their own Proof of Stake algorithm called “Slasher,” which is intended to penalize miners who would normally be incentivized to mine on both a main chain and a forked chain should a fork occur. Given that Ethereum is still pre-launch at the time of this writing, it remains to be seen whether or not Slasher or a variant thereof will make it into the final product. The core devs have stated that they intend to hold Proof of Work competitions to discover the most sustainable and “fair” way of securing the network and distributing Ether (the native cryptocurrency on the Ethereum network).
Proof of Burn is a method of currency distribution whereby miners actually destroy existing currency units to gain a chance at being recipients of the next block. The idea is to do something which is resource-intensive without actually using “real world” resources (contrasted with Proof of Work which is resource intensive while actually using “real world” resources, namely lots of energy resources).
Counterparty is an ambitious project which seeks to compete with Mastercoin as the dominant “metacoin” protocol on the Bitcoin block chain. Unlike Mastercoin, however, rather than accept bitcoins as investment which can be spent on developing the protocol, Counterparty developers instructed those who wanted to pre-purchase the available supply of Counterparty cryptocurrency to burn their bitcoins by sending them to an unspendable address (a Bitcoin address for which there are no private keys). The native currency, XCP, was then distributed proportionally to those who burned bitcoins in the allotted time. This is more of an interesting way of distributing the cryptocurrency than an alternative to Proof-of-Work mining, since the Counterparty protocol is piggy-backing on Bitcoin’s mining network, but noteworthy because it is the first instance of a 100% Proof-of-Burn distribution model and doesn’t need an independent mining network.
Levelcoin was a conceptual cryptocurrency project in the Bitcointalk forums which claimed to offer a Proof of Stake/ Proof of Burn hybrid coin but turned out to be a scam (read the thread to see the history. Lesson: don’t trust blindly). The idea behind Levelcoin was that there would be two coins: Level 1 (LV1) and Level 2 (LV2). LV1 coins would initially be created in a fundraiser, and then more would be mined over time using Proof of Stake at a rate of 4 LV1 coins per week for every 1000 LV1 coins a miner owns. To obtain LV2 coins, which each generate 20 LV1 coins per week and cannot be spent, 1000 LV1 coins would be burned. So while the existence of LV2 coins would cause 5x weekly inflation of LV1 coins, the system would require a heavy deflation of LV1 coins for each LV2 coin created.
Alternatives to pure Proof of Work are compelling to me for the same reason they are to others: burning electricity simply for the sake of burning electricity seems awfully wasteful. Solarcoin is interesting in that it’s attempting to incentivize solar energy usage which is seen as more environmentally sustainable, but is that enough? Solar panels can be quite dirty, and high-energy-consumption scrypt mining systems could still be running on that project for over 20 years. If we can drastically reduce this energy usage while maintaining a secure protocol for distributed consensus as is possible with Ripple or low-energy mining proofs, why not? Whether one of the solutions described above becomes the dominant distributed consensus protocol remains to be seen; until then, expect to see a lot of experimentation that tries to solve the problem of cryptocurrency sustainability.