Originally published at: Understand Cryptocurrency, but Don’t Invest in It - TidBITS
Cryptocurrency is volatile, expensive to trade, illiquid, and rife with scams and account hijacking. Yet it contains kernels of technology and principles that likely will dominate financial markets in the future.
Originally published at: Understand Cryptocurrency, but Don’t Invest in It - TidBITS
I think I agree with everything written (that I read). I didn’t make it half-way because it doesn’t directly involve me although I have heard plenty of recommendations to invest in it, even if they’ve tailed off recently. Cryptocurrency has been around a fairly long time now and the article confirms it still has a long way to go. The only part of it that interests me, when I want to finally spend the time, is understanding the cryptographic and technical aspects. However, I do not even know how our usual financial system works in that regard. I’m just glad it works as well as it does (which is quite good) given that Murphy’s Law otherwise applies to security on the internet.
A group of people (miners) voluntarily compete over short periods of time, about ten minutes, to solve a mathematical needle-in-the-haystack puzzle. The miner who solves the puzzle first is the winner and receives a chunk of Bitcoin.
There’s plenty I don’t understand about cryptocurrency. I can’t even get past this first step. Who issues the puzzles?
Also, I’m wondering why cryptocurrency companies are suddenly buying Super Bowl ads and sponsoring MLB umpires. Are these just well-funded ponzi schemes?
As a Bitcoin (or other cryptocurrency) miner, you run the Bitcoin software (or software for the coin you’re going to mine). The “puzzle” is built into the operation of that software.
That’s not a controversial subject at all Ignoring that, you’ll find that many of the companies provide services to people in the crypto community. I’m sure you’ve heard the adage that in a gold rush, the ones that make money aren’t the gold miners, but those selling supplies to the gold miners. There’s an element of that in crypto. Crypto exchanges (where people buy and sell cryptocurrency) take their percentage regardless of whether the people using the exchange make or lose money.
I live in an old gold mining town in California so I’m well aware of that concept. I have an old friend who’s convinced that he can make a million dollars selling NFT art even though he has no idea how it works. I keep trying to explain that adage to him.
This is the so-called proof-of-work system.
In order to add a block to Bitcoin’s blockchain (and thus permanently record the transaction(s) contained within), some amount of cryptographic processing is required. The nature of the processing is that it is computationally expensive to add a new block, but very inexpensive to verify that a block (once added) is valid.
The algorithms are dynamically adjusted by the network in order to ensure that it is always computationally expensive - so the amount of work required goes up over time as hardware gets more powerful. It is also adjusted based on network load, in order to ensure that transactions are processed in a timely manner - when there are a lot of transactions to process, the amount of work goes down, and where there are fewer transactions, the amount of work goes up.
The idea is that you can’t add a block without doing all the work. This prevents attackers from spamming the blockchain - they would require as much processing power as the rest of the network combined in order to have a significant impact.
As a reward for donating the CPU horsepower to the network, when a block is added to the network, the owner of the computer (or pool of computers) that completed the computation first is awarded a small amount of Bitcoin. This is why “miners” are willing to buy so much hardware and consume so much electricity to perform these computations.
The interesting thing is that the purpose of all this computation is not to secure the blockchain, but to prevent small players without the necessary investment in hardware from being able to directly alter (and therefore potentially corrupt) the distributed database.
There are other techniques used by other cryptocurrency networks that don’t involve massive server pools (and the energy consumed by them). One (which Etherium plans to switch over to at some point in the future) is proof-of-stake, where priority is given to those nodes that manage the most assets on the network.
There are many other proof-of-whatever systems that try to avoid network abuse by restricting transaction processing to nodes that contribute large amounts of memory/storage (proof-of-space), nodes that are pre-approved and trusted (proof-of-authority) and other algorithms.
Different algorithms are more suitable for some blockchain applications than others.
For instance, a blockchain run by a consortium of electronic parts manufacturers for tracking global supply chains, or one run by municipal governments to track land ownership would probably be best served by a proof-of-authority system, since you would only want authorized users to be able to modify the data. But a cryptocurrency designed to be independent of any central authority (e.g. Bitcoin) would definitely not want that algorithm.
Interesting point that we didn’t insert—maybe should have noted right above this, @ace @jcenters, that cryptocurrencies are founded on a white paper. Bitcoin was the model for this. First, you write a description of how your cryptocurrency works. Then you (and possibly others) implement it in code. That description in code is fixed unless 95% of mining capacity “agrees” to adopt updates that change how nodes process transactions or blocks work. (Smaller improvements or bug fixes that don’t introduce changes in block creation can be rolled out by nodes agreeing to update.)
So the Bitcoin algorithm as described in its white paper and written in code specifies the puzzle. The puzzle is basically:
- Take a set of transactions from the active pool.
- Perform a series of cryptographic operations on them.
- Add your address as a miner.
- Add a counter starting with 0.
- Run that entire chunk of text, the source of the block, through a hashing algorithm.
The puzzle is: does the resulting hash in binary have a certain number of zeroes at the start before you hit a one? Because the hash’s output can’t be guessed from the input—that is the nature of a hash—the only way to “solve” the puzzle is to use brute force. So if the first test fails, increment the counter and try again. This might have quadrillions of times before any miner solves the puzzle.
Difficulty increases by requiring more leading zeroes and decrease the opposite way.
One thing I didn’t get into is that there are really four divisions in cryptocurrency: the miners, the node operators, the participants (people posting transactions), and the programmers. The programmers may work for mining firms or be entirely independent. They could work for foundations and nonprofits.
In order to get major changes in the underlying code approved, programmers have to implement it and then 95% of blocks mined in a certain period of time have to signal adoption of the updated protocol. If they don’t, the change isn’t made. Sometimes developers will “go on strike” and not roll out other improvements needed if miners don’t adopt protocol changes. It’s an interesting tension as you would think the miners had all the leverage!
They’re well-cashed-out Ponzi schemes. People who got in early can cash out holdings to later entries via exchanges which exists to swap cash and cryptocurrency. Some people are true believers and own riches only in code. Others, wiser, diversify and get real government currency they can spend directly. The more they advertise, the more people buy in, raising the exchange price, allow promoters to cash out more of their virtual currency into real cash.
This isn’t just limited to ads during the Super Bowl and other big sporting events, crypto companies have been furiously bidding to license their names on stadiums, a few have already been successful. Major League Baseball, other leagues and teams are actively courted by Cryptos as well, and not just for stadiums, but for stuff like patches on uniforms, logos on beverage and food containers, etc.
Here’s just a few examples:
Sports attracts huge audiences, whether broadcast, online, in stadium, etc. Establishing brand recognition, expanding bases of clients, building revenue. They are also focusing on luring away traditional banking, stock trading and other financial customers, as well as spreading the word among emerging markets of potential customers. And legalized sports betting has been spreading around the US; it recently became available in New York and we are being inundated with broadcast ads for online betting as well as crypto betting.
Even though it’s months away, negotiations about next year’s Super Bowl are underway. I suspect there will be even more crypto ads on next year’s Super Bowl and most other major sports.
Another important takeaway from this article is that there is a difference between cryptocurrency and the blockchain technologies that underlie it. Although all current cryptocurrencies are built over blockchain technologies, blockchain has many other uses, which I think will prove to be far more important over time.
(TL;DR: Blockchain: Yes. Cryptocurrency: No.)
IBM is one company that is conducting massive amounts of blockchain R&D. Their Hyperledger Fabric family of products is using blockhain technologies to manage enterprise data resources over a decentralized network of (presumably IBM cloud-based) servers, and is supposed to be lower cost than doing it the traditional way (e.g. with SQL databases replicated to many locations).
Most blockchains of this form will never see life outside of the company that’s running them. Definitely not cryptocurrency. But if the products live up to the marketing hype, they may prove to be a big cost saving to companies big enough to be able to take advantage.
Another example I’ve read about (but not yet seen in practice) is for things like real estate transactions. By keeping deeds in a blockchain (run by a local, county or state government), it should be possible to perform transactions much faster and cheaper than the way it’s done today. A seller and buyer could provide their data to this blockchain and transfer the property (after a valid sale, of course) without the time consuming and expensive process of manually recording the deed at a town clerk’s office. And if multiple municipalities decide to share a single blockchain (or link them in some other way), this could allow much easier cross-jurisdictional sales.
Again, this is not cryptocurrency. A real estate blockchain like this would almost certainly be run by the existing government authorities that record deeds today, and those authorities would be the only ones with permission to execute a transaction. But even this way, it could dramatically simplify and reduce costs for these transactions.
Thank you for highlighting that! I’m not sure I did enough. I think of it as “value” versus “immutable public record” or at least “immutable record.”
Yes! Any kind of serial transaction in which records should never be deleted, but only updated or amended. If you look at county property records in the United States, they’re basically an unsecured “immutable” ledger. This would also be incredibly valuable for art provenance.
I miss the gold standard.
After a bit more reading, I realized that Hyperledger is not an IBM project, but is an open source project managed by the Linux Foundation: https://www.hyperledger.org/. IBM does sell products based on it. Many other big companies are members of this group, including
- American Express
- CVS Health
- Deutsche Telekom / T Mobile
- JP Morgan Case
Some of the Hyperledger Foundation’s more interesting case studies (at least to me) include:
- Digital transformation of Abu Dhabi’s land registry (Tech Mahindra)
- Freight invoice management (DLT Labs & WalMart)
- Online parts marketplace (Honeywell Aerospace)
- Food supply chain transparency (WalMart)
Yep. My personal opinion is that cryptocurrency and NFTs are just one big scam and they will eventually collapse into themselves because there is no intrinsic value in their concept. The tech has uses.
.but money it ain’t.
I always recommend reading Matt Levine at Bloomberg. You can subscribe to his free newsletter there, Money Talks. In today’s installment, he makes a funny and accurate summary of a big aspect of cryptocurrency:
…the investment opportunity [in web3 crypto] is … frequently … you know … a Ponzi scheme? In the sense that the returns come from more people joining the project, rather than from any fundamental purpose that the project serves. The underlying thing-ness is besides the point; what matters is setting up the right sort of token rewards to attract people to the platform.
A deep theory of a lot of crypto projects is that you can use Ponzi economics to build a real business, that you can start with entirely invented token rewards and end with a real business. I am not sure that’s always wrong, and it’s certainly interesting. A lot of crypto projects do a lot of fancy engineering with only the wispiest of materials, and how can you not enjoy that?
I’ll second this recommendation. While you probably need to know a little bit about finance and investing, Matt does a great job of explaining some fairly esoteric things. He’s also a really funny writer. I mean, anyone that can come up with “the Elon Musk Division of the SEC”…
Great article. I’m still skeptical of the claims that it will be useful in the future.
keeping identity secret without exaggerations about anonymity or privacy… Instead of trusting software developers and cryptocurrency creators—while claiming you trust no one—we would trust other entities that can be better vetted.
Isn’t that already what Credit Card companies are?
Imagine a value-transfer system that was far cheaper than the credit card system, which averages around 3% in the US
It’s easy to imagine, but that’s the amount that the market has been willing to pay, so I see no reason why cryptocurrencies wouldn’t end up at the same market rate.
Because a tightly controlled secret is at the heart of all cryptocurrencies, the possibility of reducing fraud and theft has a high upside for consumers, banks, and businesses.
A PIN is a tightly controlled secret. In the US, credit card companies have not even required us to use a PIN to facilitate a transaction. The prospect of loss of spending in their system is apparently worse than the fraud that would be prevented by a tightly controlled secret.
Fourth, a government-regulated cryptocurrency could make trans-border commerce easier without reducing accountability, fraud resistance, and protections against funding terrorism, laundering money, and engaging in financial crime. It’s typically slow and costly to send money between countries
The reason the fees are high is because that is a lot of work. Government could of course take on all of that work, but why would they? It’s expensive to do all of that. It would require a lot of tax money. It’s far cheaper for the government to simply require private companies that want to facilitate trans-border commerce to do those things. Why should the tax-payers subsidize trans-border commerce?
a future cryptocurrency could allow an organizational structure in which your stake gives you proportional voting rights in projects that aren’t publicly traded corporations. (Companies tend to have concentrated ownership and even separate classes of voting shares that dilute the effect of individual and small owners.)
Companies tend to do this because that’s what they want. Nothing prevents anybody from setting up an organizational structure with proportional voting rights. If it isn’t happening now, it won’t happen just because of some new cryptocurrency. The fact that DAOs exist in a simplistic form just means they haven’t gotten complex enough to do what people actually want—separate classes of voting shares, etc.
Ok. That I understand. There was an Advent of Code puzzle in 2016 that was basically this. I had no idea I was learning how to be a bitcoin miner!
Fidelity is jumping head first into the crypto waters:
…and the pool has no water in it.
It’s possible their announcement is premature:
In March, the Labor Department advised 401(k) fiduciaries to “exercise extreme care before they consider adding a cryptocurrency option to a 401(k) plan’s investment menu.” The Employee Retirement Income Security Act of 1974, or ERISA, requires retirement plan fiduciaries, including trustees, administrators and investment committees, to “diversify the plan’s investments in order to minimize the risk of large losses.”
The Labor Department isn’t the only U.S. regulator to issue warnings for investors about cryptocurrencies. The U.S. Securities and Exchange Commission has also cited concerns over fraud, manipulation and investor safety in rejecting several applications to approve Bitcoin spot exchange-traded funds.