A few years ago I read (and reviewed) the ORA book "Blockchain". I found it to be muddled in the way the literature of cults often is. Still, cryptocurrencies are an interesting technology and I wanted to understand them better so I tried to read a better book on the topic. That book, also from 2015, is "The Age Of Cryptocurrency: How Bitcoin And Digital Money Are Challenging The Global Economic Order" by Paul Vigna and Michael J. Casey. While this book was a serious improvement over the other one, it was still a little too oriented toward the general reader for my purposes. Still it did provide enough technical details for me to persist.

The book rightly spent quite a bit of ink unpacking just exactly what "money" is. The answer is surprisingly complicated. Keeping it simple, I was convinced that fiat money enjoys no inherent magical superiority to cryptocurrencies in theory other than tradition. The inverse of that however, does seem plausible—cryptocurrencies have compelling tricks that traditional fiat currencies can’t match.

I came away with a better, if still annoyingly incomplete, understanding of Bitcoin. I think the concept as established by Satoshi, the mysterious creator of Bitcoin, is a solid and commendable piece of technology integration. Its first mover advantage seems about offset by the design flaws that could only have been discovered in practice. But there are definitely some problems.

One thing I was fascinated to learn about was the 51% attack where an entity which controls more than 50% of Bitcoin traffic can spuriously "fork" the transaction ledger and start double spending. This seems almost an absurd problem but it is a lot more plausible than someone, say, controlling 50% of all dollar transactions. It turns out there are quite a few plausible scenarios where this problem could manifest itself. First a powerful and technically savvy nation state could feel that Bitcoin is a threat to their fiat currency; if they were, for example, extremely populous and authoritarian, this is not a completely unfeasible attack vector. Also there are many points in the Bitcoin ecosystem that are bottlenecks of power. For example, there are the exchanges which could turn malicious. The Mt. Gox Bitcoin exchange fiasco that almost took down the whole system is a flagrant example. In addition to many people aggregating to relatively few actors who actually interact with the blockchain, another issue is mining pools. Since mining for Bitcoins is a bit like playing the lottery, people join pools so that if any winners win in the club, all the participants will get rewarded. This evens out the incentive for mining. But if a certain mining pool becomes very powerful (which has happened), this can destabilize the whole system.

Speaking of mining, I learned that the mining was actually a way to pay people for verifying transaction integrity. This is a very clever design idea, but tuning it seems quite tricky. And even if it were properly tuned, does it scale? Currently there are incentives to burn through huge amounts of energy to mine Bitcoin. This may not be optimal. To be fair it’s probably conceptually no different than the avarice behind dirty gold mining.

Another scaling problem is that the Bitcoin network can currently process about seven transactions per second. This is considerably fewer orders of magnitude than Visa’s ten thousand per second. Also it takes on the order of minutes (I think about 12) to "clear" a transaction as established in the valid blockchain.

Most treatments of Bitcoin are afraid that the two technical cornerstones of the system will be the hardest to explain. For me, it’s frustrating because hashing hashes and using asymmetric cryptography are the parts that I know quite well (my interest in Bitcoin comes from this technical expertise, not to it).

Even with scant details provided by the book, one unmentioned problem I noticed that seems important is the inexorable growth of the Blockchain. People just talk about this Blockchain as if it were magically storing all this transaction information for free. Well, no. The miners can verify integrity, but surely it is clumsy to carry around with you a record of every transaction ever made by anyone ever in history. Does that really scale to billions of every day users for centuries? Let’s just say that I’ll wait and see about that.

Is Bitcoin (or other cryptocurrencies) a good idea for getting rich quick? Might be, might not be. So in other words, for risk averse people, no. But my interest comes from some of the other things advocates of blockchain based technologies promote. For example, if there is a truly immutable public ledger of all transactions (which I believe is true based on first principles and its apparent implementation) it should be possible to encode all kinds of things in the transactions. My main interest is finding some very good way to attest that some information was known at or before a certain time. So if I write a song and sing it privately at a dinner party and then months later I hear the same song on the radio, how can I show that I actually wrote that song? The answer is to hash the song’s data and put that hash into the Bitcoin blockchain. Now there is a record backed by a lot of strongly financially motivated actors to be true and accurate that I must have been in control of this song before a certain date.

The book delves into the disruption this kind of thing conceptually could cause. The book says that the BLS says that 10e6 people are employed in securities, real estate, insurance and other jobs which basically exist to provide trusted financial transactions. In theory, with a technology like cryptocurrency blockchain ledgers, these jobs are no longer necessary. You hear about the 1.8e6 truck drivers facing job loss to autonomous vehicles, but you don’t hear about the bankers who keep the financial system honest. This is all very interesting, but I’m not going to bet on the current system to quickly collapse just yet. (If I’m wrong, great!) If using cryptocurrency was the smart thing to do and people were inclined to do what was smart, then we would all be using pubic key cryptography to sign every thing but obviously most people don’t even know what I’m talking about. And that’s the same exact problem for Bitcoin.

Bitcoin could take over the dollar’s position as the world’s strong currency. Or it could collapse into historical curiosity. Or it could become a collectable curiosity like Roman coins. Or it could be taken over by some gamers or some weird application that has the perfect incentive to keep it alive. I think overall though that it is part of a much bigger change in how we do things. The printing press and paper currency were pretty huge changes that radically changed things when we look back at a graph of the centuries. But keep in mind that it was almost exactly 300 years between Gutenberg’s press and the first public library. Many of the ideas behind cryptocurrencies are sound (trust through public key cryptography, hashes are difficult to forge), but it could be centuries before we’re all properly reaping the benefits of these computer science miracles.