Anchor coins vs. altcoins
Posted: Sat May 13, 2017 10:11 am
Last week marked a major milestone for our project. It was the time that we ceased using bitcoins as our reserve currency.
When we wrote our first line of code on December 23, 2013, we developed the system with a primary goal of minimizing risk due to coin volatility. While we mine many coins, we decided to leave speculation on them to the traders, and we opted to avoid holding anything except the money owed to our customers, and our own bitcoin to be sold to dollars. Until recently, this strategy worked well: we would mine litecoins, DASH, and about 350 coins, buy the coins users wanted, sell our profit to bitcoins, and get rid of the bitcoins to dollars periodically.
As time went on, however, our risk tolerance for holding bitcoins was forced upward against our wishes. While our accounting continued to count on immediate bitcoin sales, sending bitcoins became expensive. Thus, we were forced to become traders. Every day, the question was "Will the price of bitcoins fall more today than the fees will cost to get rid of them?" We are not interested in analyzing charts just to determine whether the fees justify taking profit today, and it takes up time that could be better spent on other tasks. While the future value of work on this project is hard to quantify, one direct, concrete example is that if we were able to save just five minutes per day looking at charts, it would eliminate the need to pay cleaners $145 for the two hours they spend every four weeks cleaning the floors and toilets.
But as it turns out, we now pay far more than $145 per month in transaction fees.
This transaction cost about $8: https://blockchain.info/tx/15624a7a04e9 ... e2018d0997. When it cost about $2 to send one of these 40 days ago, I predicted that they would cost about $50 by the end of the year, and the fee costs are increasing right in line with that outcome. Had we taken no action, the cost in fees to us would exceed $15,000 this year alone, which is unacceptable.
The opportunity came when Coinbase added litecoins. Because of the API compatibility, with a few minor changes, we now use litecoins for all of our profit. Instead of collecting bitcoins, we now hold litecoins, and when we get lucky, we send excess blocks over to Coinbase for sale. Even when we don't mine the transactions that go to Coinbase, I've never seen them cost more than 1 cent and they have always been included in the next block. It is also marvelous to be able to stay at the computer for two minutes, see the money arrive, execute the sale and ACH transfer, and be done, rather than seeing the bitcoin price fall and losing $1000 while you eat dinner waiting for the money to finally arrive.
In fact, the only time we deal in bitcoins now is to pay out customers or to pay people who only want bitcoins. Last week, I needed to pay miningrigrentals, a cloud mining service, to obtain test hashpower to try to break our system. They only accepted bitcoins, so I paid them $3 with a fee of $1.20. Then, I wasted time with video games for 2 hours while I waited for them to confirm before I could actually use the hashpower for testing.
The rise in fees has caused emergent behavior in our software, as we found that our trade volume on exchanges that do not have LTC base pairs has declined. One of the strengths LTC has always had was that most exchanges offered trading in LTC base pairs, and the lower value of LTC allows coins that would otherwise be worthless to be traded. Now, LTC base pairs allow people to bypass the bitcoin network altogether and avoid its fees. Our volume at exchanges that offer only BTC base pairs, like Yobit, has declined to less than 0.2 per day, because the system considers losses from withdrawal fees when making trades in order to get the best price.
Recent developments have prompted us, and surely many others, to reevaluate the meaning of the term "altcoin." In the past, there was Bitcoin, and then there were all sorts of other currencies that were "altcoins," implying that they were lesser networks. At the time, that was true; while ETH had its smart contracts, ETH was rarely used for contract purposes, and the only advantage litecoins had over bitcoins were their extreme stability and usefulness as a store of value in comparison to the volatile bitcoin market.
Now, not only is it unclear whether bitcoins have any advantages over these other coins, a better way of phrasing it is more that it's unclear why anyone would want to use bitcoins at all with these superior alternatives available, except for legacy support.
Therefore, instead of treating Bitcoin as a category by itself, we should be referring to coins as "anchor coins" and "altcoins." Litecoin is an anchor coin for the scrypt market, ETH is an anchor coin for Dagger-Hashimoto, DASH is the anchor coin for X11, and Bitcoin is the anchor coin for SHA-256. Each of these coins is the dominant coin in its market, and the terminology makes sense because the "altcoins" in each market are almost always forks of the anchor coin. The other advantage of the term "anchor coin" is that each coin exists because it excels at a particular purpose over the others. Litecoin, with its wide acceptance across exchanges, its API compatibility, and its cheap fees, is the best coin for transacting money, while Novacoin is an altcoin based on LTC. Ethereum is the leader in smart contracts (but is hindered by a lack of base pair exchange support for value transfer), while ETC is an altcoin based on ETH. DASH is emerging as the standard for anonymous transactions, and Centurion is an altcoin based on DASH. Bitcoin has the greatest legacy support for those willing to pay for it. While coins will rise and fall, it is unlikely that any one coin will become dominant, at least in the next five years.
This taxonomy more accurately describes the current and future state of cryptocurrency. Had Bitcoin Unlimited activated a year ago, it's likely that most of these coins would not have gained traction, but since it did not, the future is now a multiple-coin world with each coin providing services to a different segment of the market. As we start to see multi-coin wallets that seamlessly change between currencies based on fees and merchant acceptance emerge, a new view is required.
It's important to differentiate between coins that have widespread support, security, and differentiating features; and ICOs and clones that are forked off these coins that often offer few new features but have high risk and volatility. Rather than the previous state of "Bitcoin" and "altcoins," we now live in a world with "anchor coins" and "altcoins." Forward-looking developers and entrepreneurs have already recognized this change, and I expect that the next few months will see a focus on adding payment support across the anchor coins and making it easy to transition between them.
When we wrote our first line of code on December 23, 2013, we developed the system with a primary goal of minimizing risk due to coin volatility. While we mine many coins, we decided to leave speculation on them to the traders, and we opted to avoid holding anything except the money owed to our customers, and our own bitcoin to be sold to dollars. Until recently, this strategy worked well: we would mine litecoins, DASH, and about 350 coins, buy the coins users wanted, sell our profit to bitcoins, and get rid of the bitcoins to dollars periodically.
As time went on, however, our risk tolerance for holding bitcoins was forced upward against our wishes. While our accounting continued to count on immediate bitcoin sales, sending bitcoins became expensive. Thus, we were forced to become traders. Every day, the question was "Will the price of bitcoins fall more today than the fees will cost to get rid of them?" We are not interested in analyzing charts just to determine whether the fees justify taking profit today, and it takes up time that could be better spent on other tasks. While the future value of work on this project is hard to quantify, one direct, concrete example is that if we were able to save just five minutes per day looking at charts, it would eliminate the need to pay cleaners $145 for the two hours they spend every four weeks cleaning the floors and toilets.
But as it turns out, we now pay far more than $145 per month in transaction fees.
This transaction cost about $8: https://blockchain.info/tx/15624a7a04e9 ... e2018d0997. When it cost about $2 to send one of these 40 days ago, I predicted that they would cost about $50 by the end of the year, and the fee costs are increasing right in line with that outcome. Had we taken no action, the cost in fees to us would exceed $15,000 this year alone, which is unacceptable.
The opportunity came when Coinbase added litecoins. Because of the API compatibility, with a few minor changes, we now use litecoins for all of our profit. Instead of collecting bitcoins, we now hold litecoins, and when we get lucky, we send excess blocks over to Coinbase for sale. Even when we don't mine the transactions that go to Coinbase, I've never seen them cost more than 1 cent and they have always been included in the next block. It is also marvelous to be able to stay at the computer for two minutes, see the money arrive, execute the sale and ACH transfer, and be done, rather than seeing the bitcoin price fall and losing $1000 while you eat dinner waiting for the money to finally arrive.
In fact, the only time we deal in bitcoins now is to pay out customers or to pay people who only want bitcoins. Last week, I needed to pay miningrigrentals, a cloud mining service, to obtain test hashpower to try to break our system. They only accepted bitcoins, so I paid them $3 with a fee of $1.20. Then, I wasted time with video games for 2 hours while I waited for them to confirm before I could actually use the hashpower for testing.
The rise in fees has caused emergent behavior in our software, as we found that our trade volume on exchanges that do not have LTC base pairs has declined. One of the strengths LTC has always had was that most exchanges offered trading in LTC base pairs, and the lower value of LTC allows coins that would otherwise be worthless to be traded. Now, LTC base pairs allow people to bypass the bitcoin network altogether and avoid its fees. Our volume at exchanges that offer only BTC base pairs, like Yobit, has declined to less than 0.2 per day, because the system considers losses from withdrawal fees when making trades in order to get the best price.
Recent developments have prompted us, and surely many others, to reevaluate the meaning of the term "altcoin." In the past, there was Bitcoin, and then there were all sorts of other currencies that were "altcoins," implying that they were lesser networks. At the time, that was true; while ETH had its smart contracts, ETH was rarely used for contract purposes, and the only advantage litecoins had over bitcoins were their extreme stability and usefulness as a store of value in comparison to the volatile bitcoin market.
Now, not only is it unclear whether bitcoins have any advantages over these other coins, a better way of phrasing it is more that it's unclear why anyone would want to use bitcoins at all with these superior alternatives available, except for legacy support.
Therefore, instead of treating Bitcoin as a category by itself, we should be referring to coins as "anchor coins" and "altcoins." Litecoin is an anchor coin for the scrypt market, ETH is an anchor coin for Dagger-Hashimoto, DASH is the anchor coin for X11, and Bitcoin is the anchor coin for SHA-256. Each of these coins is the dominant coin in its market, and the terminology makes sense because the "altcoins" in each market are almost always forks of the anchor coin. The other advantage of the term "anchor coin" is that each coin exists because it excels at a particular purpose over the others. Litecoin, with its wide acceptance across exchanges, its API compatibility, and its cheap fees, is the best coin for transacting money, while Novacoin is an altcoin based on LTC. Ethereum is the leader in smart contracts (but is hindered by a lack of base pair exchange support for value transfer), while ETC is an altcoin based on ETH. DASH is emerging as the standard for anonymous transactions, and Centurion is an altcoin based on DASH. Bitcoin has the greatest legacy support for those willing to pay for it. While coins will rise and fall, it is unlikely that any one coin will become dominant, at least in the next five years.
This taxonomy more accurately describes the current and future state of cryptocurrency. Had Bitcoin Unlimited activated a year ago, it's likely that most of these coins would not have gained traction, but since it did not, the future is now a multiple-coin world with each coin providing services to a different segment of the market. As we start to see multi-coin wallets that seamlessly change between currencies based on fees and merchant acceptance emerge, a new view is required.
It's important to differentiate between coins that have widespread support, security, and differentiating features; and ICOs and clones that are forked off these coins that often offer few new features but have high risk and volatility. Rather than the previous state of "Bitcoin" and "altcoins," we now live in a world with "anchor coins" and "altcoins." Forward-looking developers and entrepreneurs have already recognized this change, and I expect that the next few months will see a focus on adding payment support across the anchor coins and making it easy to transition between them.