Dogecoin worthlessness
This post originally appeared on reddit. I got some amazing feedback and insightful comments, which I thought warranted a follow up article. So here we go! Last time when I wrote, I mentioned the dogecoin block schedule and pointed out that the rate at which we are issuing DOGE is such that most of the second and third generations ASICs which are being manufactured targeting Scrypt coins will never see "high" returns per block simply due to the impending halvening s.
An open question for cryptocurrencies is where do we derive valuation from? The Bitcoin crowd looks to Austrian economics and the fact that Bitcoin is asymptotically finite resource as the sources of their valuation. Now I'll confess that I'm a skeptic of both Bitcoin and Dogecoin as they stand today. As a computer scientist, Bitcoin is fascinating because the blockchain and proof of work which Satoshi introduced is a viable solution to the Two Generals Problem.
This article gives a good summary of the significance of this discovery. So what does this mean? It means that the Bitcoin protocol which underlies Bitcoin, Dogecoin and all the others represents a fundamental advance in the design of distributed systems. That this advance is used to implement a distributed, verifiable "value" coin store or cryptocurrency is just the first of many problems which I expect to see revolutionized by cryptocurrency derived software. Namecoin is a fascinating example of the wider value of the blockchain technology.
So the blockchain itself as a technology is interesting. What does this actually mean for cryptocurrencies? It means that cryptocurrencies are potentially the ultimate value refuge. Another reason to hope for the value of the coins is that they provide more secure transaction capabilities than credit cards do.
As we were reminded with the Target hack, in order to process one credit card transactions, companies who accept credit cards need enough information to process an arbitrary number of arbitrary sized trasactions. The cryptocurrencies do much better at this, as it is impossible for a payment recipient to "charge" repeatedly.
Thanks to multiple signature transactions we've even solved the escrow problem, although escrow providing payment processors for coin based purchases still need to mature.
The last significant source of value I see is the low transaction costs. This means that as we see with DOGE the community can at near zero marginal cost donate to causes anywhere in the world, further increasing the potential impact and reach of previously centralized crowdfunding efforts. Now there's a dark underside to the blockchain techology: The blockchain only functions because any entity who wishes to intern records in the blockchain must perform a significant and fundimentally worthless computation in order to prove their honesty.
Bitcoin uses the sha hash function, we use the Scrypt hash function. Some coins, such as Primecoin have taken an interesting approach to the fundamental worthlessness of the shasum and Scrypt computations by trying to use scientific computing problems as proofs of work. Primecoin is in an interesting place, because verifying primality of a number or number sequence is comparatively easy. Other proposals especially for protein folding based coins, are ultimately doomed because the difficulty of verifying a solution to some folding problem is equivalent to solving it in the first place.
This means that verifying blocks and the transactions encoded therein is exceedingly slow and hard, which decreases the micropayment utility of the coin. However, mining is ultimately a network mechanism for securely processing transactions. It has a cost and no value. The per block reward mechanism first used in Bitcoin is a mechanism for purchasing the mining power required to securely process transactions by inflating the coin and expecting that miners will speculate against the future value of the coin to justify their mining costs.
Note that this is fundimentally circular. The initial miners are highly rewarded for supporting the network at its most vulneralbe, with miners who join only after the coin has achieved some modicum of stability being less strongly rewarded. Using data from http: It means that in order to be secure as a coin, our hashrate needs to increase significantly. We either need to pay more for mining, or we need to decrease our dependance on mining more on that later.
At current market prices satoshi as of this writing that pegs our per block reward at 0. So at our current network hashrate of a In comparison, mining Litecoin pays out at 0. As our return is lower than LTCs and new hyped alts shitcoins are even higher, I think it's clear that we are unlikely to see Wafflepool or other significant sources of hashing power come back without a price jump into the s at least.
Now I can sit here and run numbers, but clearly DOGE while featuring an awesome community is not at present a high ROI coin which means we are not accumulating the hashing power needed to secure ourselves against potential machinations of present whale miners ignoring the potential threat of future ASICs.
However he's the good shibe and I'm the math shibe. Assuming that error due to shibes rocking hundred GPU rigs and shibes who CPU mine average out, I think this number makes it clear that as a community we can't raise the hashing power that we "need" to armor ourselves against the potential machinations of whales and pools.
As we can't raise the raw hashing power that's required to maintain our beloved coin and our market valuation continues to fall, it seems to me that there are two plans of action.
The first is to try to raise awareness of Dogecoin through publicity stunts like Doge4water and Dogecar. I hope that these both prove to have been positive forces in the long run. Ultimately however, Dogecoin is only as useful as Bitcoin or Litecoin is if we can only tip each other, donate and hoard.
Payment processor adoption and getting businesses to start accepting cryptocurrencies are the real way to secure and increase our long term valuation.
In the short term we need to secure ourselves as well so we can get to the long term. Clearly as a community we don't have the kind of hashing power we need to do it ourselves with the "classical" proof of work structure, so I suggest that we look elsewhere, especially to proof of stake. Proof of stake is an alternative transaction verification scheme under which "miners" mine not on the basis of how much computational power they have, but on the basis of how many coins they control.
By adopting proof of stake, we would make it much more difficult for a wild whale to crush us whether out of malice or by accident. However most importantly we increase our own perceived security and stability which clears the way for more adoption down the road. I look forwards to seeing what you think of this piece. Frankly I ran the numbers as I wrote it, and I must say I'm saddened by coming to a largely negative result.
If there's interest, I'll try and do another analysis piece speculating on the price of DOGE and our required hashrate through halvenings, but that's for another day. The blog of Reid McKenzie. Dogecoin part II 25 Apr This post originally appeared on reddit. From last time Last time when I wrote, I mentioned the dogecoin block schedule and pointed out that the rate at which we are issuing DOGE is such that most of the second and third generations ASICs which are being manufactured targeting Scrypt coins will never see "high" returns per block simply due to the impending halvening s.
On valuation An open question for cryptocurrencies is where do we derive valuation from? In the shadow of the moon Now there's a dark underside to the blockchain techology: Adjusting launch window As we can't raise the raw hashing power that's required to maintain our beloved coin and our market valuation continues to fall, it seems to me that there are two plans of action. Afterward I look forwards to seeing what you think of this piece.
Social Networking Tell me what you thought! I love getting commentary good or bad. Related Posts Shelving; Building a Datalog for fun!