1. 26
  1.  

  2. 9

    Mickens’ five observations are true enough with respect to a huge range of human economic activity, but his thesis is basically that blockchains are bad because they’re inefficient ways to do computation - having to cryptographically guard against defection by participants in a distributed computation is inefficient because you can just have the legal system punish defection. But the places where blockchains are a good idea are exactly those where an out-of-band legal relationship between non-anonymous transacting parties is unavailable or undesirable. They’re another way to solve the problem that the non-blockchain economy solves by having the possibility of Visa suing you in an American court of law, explicitly meant for people who can’t or don’t want to trust Visa and the American court system. He didn’t address that line of argumentation at all, but just assumed that the legal system was clearly a better alternative for anything you might be thinking about a blockchain for.

    1. 5

      I agree with the critique. Even though I am a blockchain grump (really just have a counter-hype instinct), throughout the talk there were a handful of moments where I was prompted to think “But that’s not necessarily true.” or “But there are implementations that have that desired property” and “But there are implementations that don’t have that undesired property.” but in light of his audience for this talk, those caveats or objections are distractions.

      While every statement of the talk is not universally applicable, it is evidently not the point of the lecture.

      Looking at the one shot of the audience, there is at least one person from Accenture, and I can only assume similar corporates paid for their folks to attend the conference so as to understand what “blockchain means for us and our clients” or some such “thought leader” nonsense. It’s HBS after all.

      I highly doubt that anything on that day was intended to speak to a creative audience intent on radically rethinking how core human interactions and structures could be augmented and improved by new technologies.

      Put another way, as far as the audience is concerned saying “the legal system was clearly a better alternative” is a almost a foregone conclusion, and they only needed to be reminded of it.

    2. 5

      This guy doesn’t get it.

      His argument goes on and on about using the existing legal system, the existing fiat currency, the existing _..

      The whole point of blockchain(-esque technologies) is to have a foundation for building the next iteration of these systems, it’s not about some sort of counterculture minority trying to look cool while defying some sort of authority institution, it’s about re-engineering the authority so that it can be transparent and efficient.

      1. 15

        transparent and efficient

        On which measure?

        If my money is stolen by fraud my bank will recover it for me or credit my account. My deposits are insured against bank failure at no cost to me. If I pay someone in good faith for a job and they don’t deliver, I have a legal framework to work within to recover my money.

        I don’t want to reinvent the financial system from first, libertarian principles.

        1. 7

          I’ll add that, on keeping or retrieving your money, the banks have been even more reliable and transparent than the coins and coin exchanges. They’re still in the dark ages. Unlike the banks, many don’t even have a sustainable business model to keep everything else going.

          1. 3

            What if the bank is the one committing the fraud(see: “The Big Short” for example)?

            1. 0

              These things are also expressible using a cryptocurrency technology, heck, normal banks are cryptocurrencies, they are networked databases (using encryption) who enumerate how much store of value each user has.

              The problem you describe just has to do with the philosophy of those who made it (or rather the resources they have, they can’t expect to play police when there is an actual police who would rather not be replaced)

            2. 6

              Attempts to reengineer the authorities from the outside is the kind of thing that tends to get met with prison (if you’re lucky) or bullets. Stable systems (like governments) do not treat sudden change as a feature.

              We’ve got a pretty long history of engineering authorities from the inside out working - slowly, and demanding constant effort, but working.

              1. 2

                I am reminded of the Max Weber quote:

                “Politics is a strong and slow boring of hard boards. It takes both passion and perspective. Certainly all historical experience confirms the truth - that man would not have attained the possible unless time and again he had reached out for the impossible.”

                I spend part of my week in a shared workspace that has many startups. I am always the blockchain grump in the room who asks “Why not use a database?” or “Isn’t that just a consensus algorithm? What does a blockchain add here that justifies the indirection and performance cost?”

                Despite that role, I think it is entirely possible that the future will be greatly based on blockchain-derived or blockchain-adjacent technologies, and that this back-and-forth is exactly how we can get there.

                There are excited idealists, dreamers who want the future now, regardless of the cost, because in their mind the benefits are so great (whether to society, or to themselves). There are grumpy nerds like me who ask annoying, detailed questions about the specifics, from both a political and technical angle, but many of us are also dreamers!

                In that framing, it is not about one “side” being right, and the other wrong. It is not about “will the changes come?” Instead, we are having a protracted conversation about the speed and method of the change, along a spectrum, and our impetus to favor one particular rate or method arise from a difference in the situations we focus on.

                Using examples from elsewhere in the thread:

                @ilmu describes an impetus to faster action because of a value in avoiding the global climate crisis and the myriad potential harms that would bring.

                The list of possible calamities triggered by disrupting certain equilibria, pushing certain global systems beyond points of no return, is countless. Daily human life could look nothing like it does today, and we could regress in major areas such as education, disease prevention and mortality after several centuries of improvement! Focusing on that, I can see how one might value a higher rate of change, even despite other costs.

                As for a slower rate of change, I agree with @nickpsecurity’s “banks have been more reliable and transparent than coins and coin exchanges”, but am not seeking to put words in his mouth, so this is my take on that stance:

                I have been in a financially precarious situation before, and would endure nightmares each week wherein my small-but-essential balance would somehow be emptied overnight (by error or theft or whatever). And that was with a regulated bank that I could reasonably trust. Observing the often-cavalier attitude of most blockchain communities, I would oppose any near-term shift to insert blockchain-based anything in the financial systems that touch my food budget and rent budget! I am certain there are others who feel the same. Focusing on that, even though I acknowledge that the future is on the blockchain, I think it is a much further future, and I am okay with that.

                1. 1

                  I am reminded of the Max Weber quote:

                  Fantastic quote, thanks for that one.

                  Despite that role, I think it is entirely possible that the future will be greatly based on blockchain-derived or blockchain-adjacent technologies, and that this back-and-forth is exactly how we can get there.

                  The bitcoin paper was a huge result in computer science. I’m sure it’s techniques are going to be used over and over - it’ll just fade into the background (eg what is now considered a slightly-complicated form lookup was called ‘AI’ from the 80s and ‘expert systems’ in the 90s).

                2. -1

                  Problem is, now we have a set of constraints (global climate crisis) that need to be met irregardless of how many bullets you have. The faster we can iterate the better.

                  1. 10

                    I can buy the argument that current governments are failing to address externalities such as climate change effectively. What I do not understand though is how cryptocurrencies are going to solve said crisis. Ignoring the environmental disaster that is proof-of-work, blockchain technology is only applicable for a limited set of properties (as Mickens mentions in this talk). I am curious what your thoughts are on how such a technology can tackle a problem as complex as climate change. To me, it feels like more of the “technology will save the world eventually” argument which still has not been fully realized.

                    1. 1

                      The climate is just the first thing to give, the problem is that our economic system is unable to express the value of certain things (because it does not model value as a relative construct, rather it has central sources of truth - banks&governments).

                      A cryprocurrency is a way for us to migrate to a new value model gracefully.

                      This would also come with the added benefit of causing global action to counter climate change, but only if the new model was effective, and I believe the migration would not happen any other way.

                      Every era has an idea and many individuals will have this idea in response to the problems of the era. The problem of this era is complexity, it is why we are helpless to deal with this climate crisis, precisely because of all our carefully tuned equilibria (that people have been defending as more mature and reliable than cryptocurrencies).

                      The idea of this era is category theory, or more precisely, compositionality, the mathematics for predicting emergent effects and taming complexity. Statebox is the only place I see actually trying to build compositional game theory into their platform so that we may build functional economies.

                      Basically I believe that the only way we can have a ‘world government’ is via technologies like this (consensus in distributed computing).

                      1. 2

                        World democracy is a laudable goal, but I don’t understand how distributed computing/cryptocurrencies deal with the real blocker on democracy: entrenched power.

                        How is cryptocurrency going to stop billionaires and enormous companies from controlling the majority of human production?

                        1. 1

                          Those entities are powerful / rich in the current value system. If that system is gradually replaced then the units of measurement within it (dollars or whatever) will depreciate in relation to the units of measure in the new system.

                          If the new system is more efficient / adaptable then it means entrenching power is more difficult.

                          I wasn’t really arguing this point though, I was pointing out the necessity of these technologies for global organization, not that they are sufficient by themselves.

              2. 4

                can i mention that bitcoin is an environmental disaster regardless of its other technical merrits/flaws? https://www.economist.com/the-economist-explains/2018/07/09/why-bitcoin-uses-so-much-energy https://www.theguardian.com/technology/2018/jan/17/bitcoin-electricity-usage-huge-climate-cryptocurrency

                edit: updated the news articles

                1. 3

                  These are articles from authors and publications with specific agendas: they have a dog in the fight with their livelihoods depending on existing economic systems not being taken over by some bitcoin-like arrangement. Here is an article from another perspective:

                  https://qz.com/1364657/bitcoins-energy-consumption-isnt-as-bad-as-you-think/

                  1. 3

                    Electricity is 90% of the cost to mine bitcoin. As such, bitcoin mining uses an exorbitant amount of power: somewhere between an estimated 30 terrawatt hours alone in 2017 alone. That’s as much electricity as it takes to power the entire nation of Ireland in one year.

                    Indeed, this is a lot, but not exorbitant. Banking consumes an estimated 100 terrawatts of power annually

                    Bitcoin’s market cap is $154 billion.

                    The market cap of the top five global banks is $1.6 trillion.

                    Banking seems to have a hell of a better ROI than Bitcoin.

                  2. 1

                    It is also a proof of concept coded by someone who has managed to stay anonymous despite global scrutiny, that exercise must have costed some cognitive bandwidth… What this tells me: We. Can. Do. Better. The sooner we do the sooner we can put down the poisoned puppy called bitcoin.

                  3. 2

                    He’s not wrong about some facets of what he says, particularly that given a real-world identity one can just rely on the existing legal system to enforce contracts. But here’s the thing: the existing legal system is really, really expensive (in the U.S., we spend about 38% of GDP on government). Is it possible to use something like a blockchain to provide many of the same benefits more cheaply and/or more accountably and/or with less susceptibility to corruption?

                    I don’t know, but it’s an interesting question. We’ll always need some form of physical government to provide physical security, but do we need a government to provide financial security?

                    1. 4

                      Is it possible to use something like a blockchain to provide many of the same benefits more cheaply and/or more accountably and/or with less susceptibility to corruption?

                      I’m a big cryptocurrency fan but I don’t think that anything like blockchains could possibly provide any government services in any sort of useful fashion. It very quickly degrades to real world identity problems that cannot be solved well without trust. If you have trust, you don’t need a blockchain.

                      1. 3

                        the existing legal system is really, really expensive (in the U.S., we spend about 38% of GDP on government)

                        Enforcing legal rules (courts and law enforcement) is a small part of that budget.

                        https://www.thebalance.com/u-s-federal-budget-breakdown-3305789

                        The discretionary budget will be $1.426 trillion. More than half goes toward military spending, including the Department of Veterans Affairs and other defense-related departments. The rest must pay for all other domestic programs. The largest are Health and Human Services, Education, and Housing and Urban Development.