[header date=”4 December 2018″]Bitcoin mining risks falling into a death spiral, do crypto futures have a future, and Algorand’s chief engineer explains the project.[/header]
The hidden force guiding bitcoin mining
One of the more maddening concepts in the cryptocurrency world is “difficulty,” the measure of how tough it is for miners to find a block and add transactions to the bitcoin blockchain. Today, finding a block—a “page” in the blockchain’s digital ledger—by solving computational problems is rewarded with 12.5 bitcoins.
By examining the time stamps of previously-produced blocks, the bitcoin network automatically adjusts its difficulty up or down to correspond to the amount of computing power, called the hash rate, mining on the network. The more miners there are, and the more hash power there is, the harder the network makes it to find a block. The fewer miners there are, and the less hash power there is, the easier the network makes it to find a block.
You might compare it to a person’s heart rate, which increases to pump blood faster during exercise and decreases to a resting pace during sleep.
The bitcoin network updates its difficulty every 2,016 blocks, or roughly every two weeks, and automatically targets an average block production time of 10 minutes.
For miners, difficulty is a constant battle. It drives them to purchase the most expensive, high-end hardware when bitcoin’s price is rising. It’s also why some miners turn off their equipment when the price plummets, because it’s no longer profitable to run the power-hungry machines. Indeed, that is why savvy miners seek devices with the highest hash rates and optimize around the lowest electricity costs.
This past week, Breaker Magazine published an overview of difficulty, suggesting that bitcoin’s adjustment mechanism is “why bitcoin will never die.” I don’t fully agree with this claim.
It’s true, bitcoin’s difficulty adjustments help keep it alive by beckoning miners with opportunities to profit when their interest might otherwise wane. But it’s not a fool-proof system. It’s conceivable that the bitcoin price could fall dramatically enough to render the network effectively dead, or close to it. This is the “death spiral”—the possibility that the network’s hash rate will plummet too fast for the difficulty to reset in time.
If the price of bitcoin falls, mining becomes less profitable, so miners turn off their machines. If enough miners turn off their machines, then there might not be sufficient hash power to find a block, so no transactions would go through on the network (or, if they did, it would be at a snail’s pace). Then, if spooked investors lose faith in the network, they dump even more bitcoin, sending the price lower, making mining even less profitable and leading to more miners walking away.
You get the idea. Of course, bitcoin trading on exchanges could continue (where transactions aren’t actually logged on the blockchain, just on the bourse’s internal records), but the network itself could quietly collapse in the background.
This might not happen today or tomorrow—it may never happen—but it’s an ever-present threat. One of bitcoin’s greatest strengths, its ability to readjust mining profitability, is also one of its greatest weaknesses. The probability of the death spiral might be low, but it’s the sort of Black Swan event that should make all crypto investors uneasy.
As bitcoin’s difficulty decreases, we should remember that the network’s robustness depends on much more than price alone. The entire ecosystem relies on these sorts of bizarre and countervailing incentive structures, which are often far more influential than individual announcements, like coin listings, which have only short-term effects. The difficulty is figuring out which incentives will prevail. —Matthew De Silva
Further reading:
An Honest Explanation of Price, Hashrate & Bitcoin Mining Network Dynamics (Medium)
What are the units of difficulty? (Bitcoin Stack Exchange)
[supplemental headline=”Market Chatter: Futures are now”]
[img src=”https://cms.qz.com/wp-content/uploads/2018/12/CME_daily_bitcoin_futures_contract_volume_Volume_chartbuilder.png”]
Nasdaq is planning a bitcoin futures contract for early 2019, joining the other big US exchanges in offering a derivative that allows traders to hedge or speculate on bitcoin’s price at some later date. Nasdaq is partnering with investment management company VanEck, which is among the parties that have been so far been stymied by US regulators in their attempt to start a bitcoin exchange-traded fund.
Launching a new derivatives contract is an uphill battle at the best of times, as most new futures fail to catch on, and it’s usually a winner-takes-all competition for the exchanges. CME and Cboe, which launched bitcoin futures trading a year ago, have a head start, and even those trading volumes are puny compared with their more popular contracts. The volume of CME’s bitcoin futures have averaged 8,000 contracts a day in recent weeks, compared with more than 1.4 million per day for a contract of West Texas Intermediate crude oil. This is despite offering programs that provide rebates to traders, essentially paying them to buy and sell bitcoin futures.
The bottom line: The crypto world isn’t pining for a new bitcoin futures derivative. This news would seem to have little impact in the near-term, but things could get more interesting down the road if US regulators approve an ETF, setting up new arbitrage opportunities for professional traders and perhaps attracting renewed interest from the retail set.
[/supplemental]
On the record: Naveed Ihsanullah
[img src=”https://cms.qz.com/wp-content/uploads/2018/12/Naveed-Ihsanullah-Algorand-engineering.jpg”]
Naveed Ihsanullah is head of engineering for Algorand, a cryptocurrency project led by MIT professor Silvio Micali. Prior to Algorand, Ihsanullah was a senior engineering manager at Mozilla.
Quartz: When you were at Mozilla, what were your thoughts on blockchain? And why did your perspective change?
Ihsanullah: Fundamentally, when the electrical costs of scaling are so high that you have cities in China parked on hydroelectric plants, it didn’t feel like a real-world solution. It felt like something else—a curiosity—but I didn’t think it was really going anywhere.
Silvio described the problem to me, and he explained his really elegant solution to this problem. I was excited by his thoughtfulness on the space, but what captivated me was the one-two punch of Silvio, covering all of the academic research aspects of blockchain-crypto, and Steve [Kokinos], who at that point was an advisor to the project. He was recently announced as Algorand’s CEO. Steve brings a real-world business acumen that most of the startups in the blockchain space don’t seem to have. They’re excited about the technology and they have great ideas, but then their implementation—the go-to-market part of implementation—always feels like something’s missing. So, I was really excited to see this company had both of these.
How would you explain Algorand to the layman?
Algorand is the first fundamental blockchain technology that addresses the blockchain trilemma, which is to simultaneously solve for scalability, security, and decentralization. [Algorand] is able to [process transactions] incredibly fast, with constant scaling effectively. The non-blockchain part is that we’re much more thoughtful about the economics of what a system needs to be a useful currency.
How does Algorand accomplish that?
The way professor Micali likes to describe it is that there’s a lottery system, the VRF [verifiable random function]. [Algorand] selects a committee that’s one-use, that makes a decision, and then moves on.
Statistically, the difference between [using the committee of computers system] and everyone voting is insignificant. And there are mathematical proofs which comfortably assure that’s the case.
Is Algorand creating its own network?
Yes. We would not be able to get the security, performance, and scalability guarantees, if we weren’t a fundamental blockchain. We are our own blockchain, with our own nodes, our own system talking to each other.
Is there going to be an Algorand coin?
Yes. There will definitely be an Algorand coin—it’s probably going to be called the “Algo.” And that will be the base currency for this ecosystem.
What’s the distribution plan for the Algo?
We’re looking at a lot of different ways to do that. We have the tension between governmental regulations and what regions we want to be legal in. I think our ideal goal would be everywhere—and to maximize decentralization. So, we don’t have a perfect answer there, but some of the more traditional mechanisms that you saw during the 2017 gold rush, like the ICO, are unlikely to be compliant in all the [target regions].
It won’t be exactly an ICO, but we’ll have some way of releasing the token. But I don’t think the traditional ICO mechanism has passed muster. It’s not been shown to pass all the hurdles yet, at least in the United States.
When’s your target launch date?
Early 2019.
What’s it been like going from Mozilla to a blockchain startup?
At Mozilla, there are a lot of excited people who want a perfect utopia for the world. The blockchain culture and universe encompasses some of that as well. I buy into some of that, but I’m much more pragmatic about how we get there. I saw that pragmatism in Algorand, and that excited me—idealism grounded in pragmatics.
Ihsanullah’s recommendations
Books: Debt: The First 5,000 Years by David Graeber, the Foundation trilogy by Isaac Asimov
Hobbies: Road biking, automating parts of his house, guitar
[mailto filter=”View from and interview feedback” subject=”Feedback”]Know somebody who we should profile? Let us know.[/mailto]
[supplemental headline=”De-jargonizer: Full Node”]
A “full node” is a device, like a laptop or desktop computer, connected to a blockchain network that stores a complete history of all the transactions that have occurred. These nodes also implement the associated network’s core consensus rules, such as standardizing the issuance of the block reward, checking transactions signatures and data formats, and preventing double spending. A system of full nodes allows the blockchain to operate as a peer-to-peer, decentralized network.
For bitcoin, each full node stores a copy of the entire record of the cryptocurrency’s usage and production. At the time of writing, the bitcoin blockchain is about 193 gigabytes.
Further reading:
Check out a live bitcoin node map (Bitnodes)
What Are Bitcoin Nodes and Why Do We Need Them? (CoinDesk)
Explain like I’m 5, what are the incentives to run a full node (Reddit)
[mailto filter=”Jargon” subject=”De-jargonize this…”]Heard a new crypto term? We can tell you what it means.[/mailto]
[/supplemental]
Please send news, tips, and futures contracts to privatekey@qz.com. If this email was forwarded to you, click here to sign up for your own subscription, which includes a free two-week trial. Today’s Private Key was written by Matthew De Silva and John Detrixhe, and edited by Oliver Staley and Jason Karaian. The proof of the pudding is in the eating.