People say that bitcoin and crypto overall are antifragile, but what does that really mean? And how does it work?
The basics
When asked to think about the opposite of fragile, most people think of robust. Like the difference between a crystal glass and a rock, one remains the same when it suffers an impact, whereas the other is damaged by even a small impact.
But the opposite of fragile has to be something that becomes stronger after the impact, otherwise it would simply be neutral like that rock.
So if the glass has a mathematical value of -1, the rock is 0 and the antifragile is +1.
Fragile – Concave, high sensibility to disorder
Antifragile – Convex, benefits from disorder
Antifragility is a mathematical property of complex systems, where there are several parts interacting with each other in a nondeterministic and nonlinear manner.
Nondeterministic means it is very hard to understand cause and effect because there are too many variables, which by itself makes comprehension difficult, since when those variables interact the possibilities multiply explosively.
My favourite example of complexity is the human brain, with 86 billion neurons on average and 10,000 times as many connections.
Somewhere within the nearly limitless interactions between your neurons is your unique self.
That is the explosive complexity that makes you, you!
Nonlinear is dose-response complexity, such as in physiology where doubling the medicine dose will not necessarily result in faster or better healing. The medicine can become poison. So as the saying goes: “the difference between medicine and poison is in the dose”
Of course the nonlinear concept is not just applicable to medicine. It is present in Metcalfe’s Law: a network’s value is proportional to the square of the number of its users.
“Networks have ‘network effects.’ Adding a new participant increases the value of the network for all existing participants. Network effects thus create a winner-takes-all dynamic. And the Rulers of these networks become the most powerful people in society.”
Antifragile systems are comprised of small parts that themselves are fragile. In fact these smaller parts, or subunits, need to be fragile to make the whole antifragile.
This happens through the feedback generated by the failures of several subunits under stressors, those failures are discarded, so the remaining survivors are able to maintain a functioning system and respond better to such stressors.
These subunits can have great variability amongst themselves like species in nature, or they can be more uniform as nodes in the Bitcoin network. Either way they respond and adapt to internal and external forces in the system.
A difference is that the existing genetic variety in a population is random and serves no purpose by itself, it just appears to have a purpose (teleonomic process) in retrospect.
Only when there is selective pressure such as changes in the environment, more predators, lack of resources, diseases and so on; that random variety becomes purposeful by adapting the system to change.
In a crypto network purpose is more or less predefined.
Both natural and crypto ecosystem benefit from The Disorder Family:
The Extended Disorder Family (or Cluster): (i) uncertainty, (ii) variability, (iii) imperfect, incomplete knowledge, (iv) chance, (v) chaos, (vi) volatility, (vii) disorder, (viii) entropy, (ix) time, (x) the unknown, (xi) randomness, (xii) turmoil, (xiii) stressor, (xiv) error, (xv) dispersion of outcomes, (xvi) unknowledge.
These systems are particularly fascinating when you realize that they usually work in a fractal manner: components of the system contain even smaller parts with the same properties as the whole. This is called fractal self-similarity and it shows interesting characteristics when we zoom out from micro to macroscale.
Maintaining our biology template, skeletal muscle cells become stronger after being exposed to the stress of an intense work out session. As far as we know tiny damages in the intracellular structure lead to stronger cells after a period of recovery.
Zooming out from cellular structure we can look at individuals in a population, in this instance epidemics are among the most striking disorder agents. Epidemics have decimated large portions of the European population throughout the centuries and yet it was that constant stress combined with enough genetic variety that produced resistant individuals, but only after generations.
The benefits of this adaptation goes to the larger population throughout time instead of all the people who suffered the chaos. This is characteristic of every scale we look in several distinct systems – subunits (individual) pay the price and the whole (collective) get the reward.
Also worth noting is how this larger physical/spacial scale is also accompanied by a larger temporal scale.
Too much stress
Eventually there is a point where the system can’t handle disorder.
When Europeans arrived in the American continent, the native population had no previous contact with those European germs. Germs also adapted to the population, densely packed cities well connected throughout the continent and a variety of domesticated animals made a perfect breeding ground of strong diseases.
The level of disorder those diseases caused in the native population was so large that led to a mortality rate of about 95%.
The amount of disorder is very important because there is a limit on how much a system can take, beyond that limit any antifragile system turns fragile.
Overall, the larger the system scale the larger the stress it can handle.
So any system can benefit from the disorder family up to a point, then after that it will fail just like any fragile system.
The payoff of a Ski resort show its income increases gradually with more snow, however, if there is too much snow the payoff curve inverts.
The ski resort is not a complex system, it just offers a simple illustration of how volatility can become detrimental if the dose is high enough.
Modeling reality
The world is a complex and complicated place, due to the massive amount of variables and the infinite possible interactions between them we can never fully experience reality as it is. We can use reason and knowledge to complement our limited perceptions of the world.
We can also create models to improve our understanding, in fact our senses are models of the world outside our minds. Sound exists only inside the neuronal symphony of our brains. The same is true for every sensory input we have. Our inescapable personal matrix.
Using reason and creating models of reality we supplement our limited perceptions, like the earth being static with the sun and stars revolving around us. Then, sooner or later, scientific knowledge acquired throughout the centuries can be transformed into useful technology.
Of course every model has limitations, as Benoit Mandelbrot said:
“All models by necessity distort reality in one way or another. A sculptor, when modeling in stone or clay, does not try to clone Nature; he highlights some things, ignores others, idealizes or abstracts some more, to achieve an effect. Different sculptors will seek different effects. Likewise, a scientist must necessarily pick and choose among various aspects of reality to incorporate into a model.”
With this in mind I want to present three models to visualize the antifragility of the crypto ecosystem. One is not better than the other, they are just different perspectives that complement each other.
Layers
This model was proposed by the economist and great spokesman of antifragility Richard Rytenband.
It consists of visualizing the crypto ecosystem expanding out from Bitcoin.
The main takeaway from this model is that the first layer, Bitcoin, is never exposed to the risk of ruin. Since the changes and updates are always very conservative and slowly carried out by developers it doesn’t get exposed to systemic risk.
All the wild experiments and new ideas are tested on the second layer of Altcoins. Alts, unlike Bitcoin, are exposed to systemic risk and can fail completely.
Any failure, while very unpleasant for investors and developers, would be beneficial to system as a whole because failures always provide new information.
That new information can be slowly incorporated into Bitcoin. Thus creating a convex first layer:
Few limited downsides and potentially unlimited upside.
This is why Altcoins are integral to the crypto ecosystem.
The experimental characteristics of the second layer help explain why Altcoins have such a larger profit potential compared to Bitcoin. The higher risk has to be compensated by larger gains, this happens organically like in any natural system, no need for central authorities dictating regulations on returns.
The second layer has fractals in base protocols like Ethereum, Waves, NEO, EOS.
So base protocols can have different experiments running on their own extra layer, just like before these extra layers serve as a new information and experimentation source.
In any project the risk of ruin always increases with more complexity, time, size and development velocity. So developers rushing to finish big projects are increasing the likelihood of sacrificing their ideas for the “greater good” with no personal payoff, remember the individual pays the price and the collective gets the reward.
Layer1 > Layer2 Altcoins (base protocols) > Layer 2 Altcoins
Before moving on to the second model let’s look at a complementary, and essential, mathematical concept to antifragility.
Lindy Effect
It is the reason that the classics never go out of style, why people still read books that are centuries old and are expected to continue on reading for centuries to come.
First described by Albert Goldman in The New Republic magazine in 1964; comedians used to gather in Lindy's delicatessen in New York. Where they noticed that the expected length of a comedian’s career was proportional to the time spent on stage or TV.
Later Benoit Mandelbrot mathematically formalized the concept and Nassim Taleb made the distinction between perishable and nonperishable:
Perishable – Fragile towards time (short theta): It ages and die.
e.g. Individual organisms
Nonperishable – It doesn’t age, it resists time but it perishes as a result of fragilities towards other disorder agents (short gamma)
e.g. The great pyramids
Nonperishable (Lindy) – It ages in reverse and it benefits from everything in the disorder family (long all)
e.g. Almost everything that is informational – genes, ideas, broad categories of technologies (more on that soon)
In essence the Lindy effect shows that anything (nonperishable) that survives one day has, on average, its life expectancy extended by another day. This is not to be taken literally but it is a good probabilistic indication of survival.
Categories
My first encounter with this model was by the great author and futurist thinker Daniel Jeffries.
Here we have to zoom out from the everyday experience we have with technology and see the broader characteristics behind them.
It’s easy to see things like those old betamax and VHS tapes, DVDs and blu-rays. The old Ford Model T and the new Tesla Model S. In the more abstract world of cyberspace examples range from MySpace and Facebook to Bitcoin and countless Altcoins.
But those are just iterations, that zoomed in view of everyday experience.
The zoomed out view is of the broad categories: video recording, cars, social networks and decentralized ledgers.
The zoomed out view throughout time will show that iterations are the small fragile parts of the technology ecosystem and categories are the antifragile whole.
Those recording mediums are all dead or dying, the Model T now is just a symbol of ingenuity and MySpace was taken over by Zuckerberg’s panopticon. Any technology iteration can die.
Categories, on the other hand, are the nonperishable-lindy survivors.
Once we visualize cryptocurrencies in their categories it becomes easy to understand how the system is evolving.
Bitcoin started the crypto revolution because it was the first decentralized solution to the double-spending problem.
Litecoin, the first altcoin, started experimenting on the same ledger architecture, then soon after there was an explosion of slightly different blockchain takes. During that crypto-cambrian explosion the first DAGs came out to experiment with a different ledger architecture (although blockchains can be considered a simplified form of DAG).
Constant evolution and experimentation are the norms in natural systems.
On the works for several years, Radix has a singular approach to seemly unbeatable trilemma between decentralization, security, and scalability. All with a new consensus and ledger architecture.
But distributed ledger technology is actually a subcategory. The basic problem it solves is how independent nodes come to agreement about a state of reality. So the complete category is Distributed Consensus Technology.
The Internet started out free and open but for years it’s increasingly becoming a walled garden.
That’s why projects like MaidSafe are aiming at creating a distributed autonomous network, like the Internet was supposed to be.
Something we urgently need.
Tree branches
And here is my slightly different take, a hybrid between both previous models.
Just like a tree of life can be used to represent biological taxonomy we can use one to represent technology as it evolves.
The trunk is the main category, the branches and leafs are the iterations.
Antifragility and the Lindy effect are stronger closer to trunk, the closer you go towards the branches and leaves the weaker it gets. To the point of becoming fragile at the edges.
I used small branches to represent Radix and MaidSafe because although both are closer to the trunk, unique in each category, they are not yet released on the real world. So neither had a chance to prove themselves before the time expert - Lindy.
The purpose of this model is to have a dynamic representation of the crypto world flowing in time. Imagine that the tree is alive and always growing, leaves and some branches can fall to fragilities but new buds appear all the time.
You can argue for a different classification and that’s fine, but here is the main takeaway:
As time goes by, backwards comparability between different projects becomes harder. Even within a very specific subcategory. It won’t always be possible to retrofit a new tech developed in the edges to another that has had time to differentiate or is closer to the base.
Zooko Wilcox talks about this at the end of this podcast.
It is as if the branches become “woodier” with time, which means achieving a useful hybridization without compromise to the stability can be a challenge with diminishing returns.
There has to be a balance between stability and backwards comparability with improvements and new features. Which means sometimes it’s better to work on a new tech rather than to improve an old one.
In the grand scheme it doesn’t matter because the trunk should remain the same, resisting all but the most drastic elements of disorder.
Conclusion
Distributed consensus technologies are not invulnerable, anything can become fragile if disorder gets to a high enough point. However this is a large tech category and that point is very high indeed.
Analogous to zooming out from small groups of taxonomy classification, from species towards life:
Antifragility and the Lindy effect gets stronger in each step.
No matter how many species go extinct life goes on.
No matter how many crypto projects fail, the crypto ecosystem is here to stay.
So assume the intertemporal view and long the Distributed Antifragile Crypto, short the centralized and fragile.
Disclaimer
Instead of the old cliché this is not financial advice... here are the five rules of science from Neil deGrasse Tyson:
(1) Question authority. No idea is true just because someone says so, including me.
(2) Think for yourself. Question yourself. Don't believe anything just because you want to. Believing something doesn't make it so.
(3) Test ideas by the evidence gained from observation and experiment. If a favorite idea fails a well-designed test, it's wrong. Get over it.
(4) Follow the evidence wherever it leads. If you have no evidence, reserve judgment.
And perhaps the most important rule of all...
(5) Remember: you could be wrong. Even the best scientists have been wrong about some things. Newton, Einstein, and every other great scientist in history -- they all made mistakes. Of course they did. They were human.
Science is a way to keep from fooling ourselves, and each other.
Special thanks to:
Richard Rytenband
Dan Jeffries
Both inspired me tremendously in this article.
Also Marco Batalha and Kady Coelho for helping with proofreading and graphics.
Thank you very much for reading!
If you like this article feel free to tip, share and subscribe.
Bitcoin - 1Ny8DefBzKXDpaBCGrFomEJ4uNxFGiJazH
DAI - 0x6314526213b16aF52BBe227d06c1e8F1e662aa00
Dash - XoCRnBqGFMjdP8PwKfF7Rrb8hnCKLbE27A
Decred - DsacQEaY9gXT6Bjvx9U1SLH3YbhHhkLBhjE
Ethereum - 0x6314526213b16aF52BBe227d06c1e8F1e662aa00
Monero - 4418L4CxTQsa6hz26bEtfEBW4iFvnRfYw6A6gq1Up6qQFLsR8NS5qyvKXc1HieV4HFP2vPF84YwEw9QWDWrnk2uRBA2yqFq
Nano - nano_1r15f1naypwn87zxq75ztx7pdx3ao9hspehh1jijjhb9c3a5jn6kc7ske9gn
NEM - ND6OZRXOWIXHTAOW4YBWDNXKGHLQWYVSU4WGGVPL
Zcash - t1fkoBKABLETDcNQg9XXxhh3ikswvhPf899
Contact - Keybase