The End of Innovation: The Problem and a sneak peak at How To Fix It
I would like to give credit to a few people first, the ideas in the essay I refined through conversation with many of my fellow collegians; A special thanks to Josh Salthouse, Will Power, Harry Isles, Charlie Strong, Campbell Rickard, Anthony James and Chris Chamberlain. And of course a thanks to all those others who have been there for me during my ups and downs haha.
There I was languishing in obscurity and boredom amongst 200 of my senior year compatriots, still barely filling a seventh of the theatre. Costa Hall to be exact, the same hall I would later receive my mathematics award in.
Impressive only in scale the theatre is extraordinarily bland, in a remarkably dystopian display, the walls, the carpet and every seat are identically coloured a kind of muddy purple brown. The lighting is entirely artificial, one is repelled by the totalising order of the place. It is the kind of thing you might imagine seeing in the nice parts of Moscow or St Petersburg in the USSR if it still existed and was in the business of building 5th rate modern Universities.
There we were being lectured from a man atop the stage with the largest PowerPoint presentation I have seen in my life. He condescended at first to excite and cajole the vain ambition of the cohort but then in a somewhat bizarre fashion — as though breaking the fourth wall on the scam being perpetrated — he started to describe how education was being acquired by everyone, that over half our generation were going to acquire a Bachelor’s Degree and how it was necessary for us to obtain a degree even if we just wanted to have 2 children and a house with a white picket fence.
It was just a short glimpse then the narrative quickly returned, rushing in to suppress any doubt. It was in fact he told us that society was so much more complicated now, that is why we needed a degree. An investigation of the facts belies his propaganda.
The data above from the ABS presents a confusing and dire image. The problem is staring us right in the face, we have an explosion of students graduating with degrees in Society and Culture, followed up by those studying Management and Commerce degrees. Meanwhile Engineering degrees fall just short of population growth (=> proportional decrease).
A look at the high-growth subcategories raises the question “Accountants?!”, what are they accounting? Software applications have dramatically reduced the need for human labour in that sector and that trend will only continue with ML solutions. Without even yet mentioning the fact that a University degree is overkill for accounting in the first place. It is more than possible with the resources available online today to teach oneself accounting for free.
We appear to live in a strange dystopia that is far from a dynamic free market society since after all industry has not caught on. HR drones from Pitcher Partners and EY did us the favour of coming to College to waste our time by telling us how they would like us to behave blah.. blah… to have the privilege of working for them at a lower rate than you would get paid as a high school mathematics teacher. They told us they did not care about WAMs so long as they were above 60 and that online degrees were fine, all they cared about was the degree. I asked “would you consider someone without a degree if they were able to demonstrate superior knowledge of the subject matter?”, “No” was of course the reply with the unsatisfying response that it was due to discipline or something (seeing as they said online was fine they had conceded it had nothing to do with social or soft skills), of course the discipline argument makes no sense considering there is nothing more self-disciplined than teaching yourself the content of an entire University degree although that sounds like more of a feat than it actually is especially for arts and commerce (I have taken some subjects).
It is remarkable how everyone in the key institutions of governance and the private sector all appear to be on a script. Notably it is always the careerists not the entrepreneurs (the real ones), that are on the script — these careerists need a better name I shall call them the nouveau científicos.
The nouveau científicos are different from previous technocrats because they emphasise the role of uncertainty and the fallibility of the human mind. They do not believe we can possibly know what the future will hold but they are sure it is good.
Given the dysfunction presented here with regards to the academy (and there are many more statistics and figures one could present until the reader collapses from exhaustion) my reaction to the oft heard phrase “dizzying pace of change” should be obvious, we are in fact experiencing a tremendous rapid slowdown in progress.
It becomes much clearer when we just take a look at the technologies. Let us compare the two halves of the last century 1920–1970 vs 1970–2020.
From 1920–1970,
By all accounts one would not imagine this period to be one of immense progress given that it transpired just after the West had been ravaged by the First World War and the world by the Spanish Influenza and the bookend on the other side was the middle of the Vietnam War. In between is sandwiched The Great Depression and World War 2. Countries leveraged themselves up to the hilt financially as the modern nation state reached the climax of mechanised destruction and yet they miraculously recovered.
This was possible because of the widespread adoption of an endless stream of life changing technologies. Where do we start? Vacuum cleaners, the motor vehicle, radio, television, refrigerators, air conditioning, the movies, antibiotics, vaccines, nuclear power, microwaves, plastics, computers, washing machines, air travel, etc. Not to mention the successful large scale public sector projects that are unimaginable today — countries building interstate highway systems, the Manhattan project, putting mankind on the moon, Arpanet.
This is why the greatest generation earned their name, they were the greatest generation and we generation Z’ers should look back to them — past the boomers and the silents if we want to put the world back on track.
From 1970–2020
The 70s in contrast to the 20s had almost every reason to succeed. Wars were trivial by comparison, little in the way of infectious diseases, after the Vietnam war a relatively content population, low wealth and income inequality and plenty of opportunity.
From the 70s onward you have very slow technological progress. Biochemistry does not live up to its expectations, all the fields of engineering begin slowing down and none of the aforementioned technologies are displaced they are only incrementally improved. The one glaring exception to the rule is information technology where almost science fiction progress has prevailed.
Personal computation, laptops, tablets, phones, video games, machine learning, the internet and cloud computing. (What about the human genome project? It has not translated to real gains yet.) Progress has become so isolated to information technology that it has become synonymous with technology itself.
There is a sense in which information technology is a kind of super technology, its ability to work with large amounts of data and carry out what for humans would be impossibly precise numerical calculations nearly instantaneously means it has the ability to have a kind of multiplicative as opposed to merely additive affect on progress. To make this more concrete what I am saying is information technology is a tool that allows you to make better tools as opposed to a washing machine which will improve your life and productivity dramatically but will not help you build better washing machines or other types of technology.
I cannot comment too heavily on the history but we can use the forge as an analogy, forges let you build better tools that in turn allow you to extract and refine higher quality ore leading again to better tools in a feedback loop, along the way the better tools also enable you to build more additive technologies such as larger more robust buildings. These kinds of super technologies will provide exponential growth for a time but they will reach saturation at a point, (so that they actually look more like an S curve, Sigmoid functions like the one shown at left are used to model other effects like population growth which naturally must saturate after a certain point) and there is evidence that this is already happening with information technology.
People have a belief that information technology is the exception and unlike the forge we will not be required to fundamentally shift our mode of production to realise exponential growth once more. This belief in part seems to be based on the idea that genius engineers will sometime soon invent a form of intelligence greater than their own — a Strong AI. In futurism circles this is known as the singularity.
In mathematics a singularity is a point at which a function is not well defined. It may converge to different values along different paths or go to infinity, etc (or as in the example at left do both, converge to positive and negative infinity at x=0).
If Strong AI were to be invented something like this would occur and we would not be able to say anything definite about what happened next but most hypotheses are either utopian or dystopian — most are dystopian. (For futurists they always seem to be utopian).
If you have done a bit of research on how modern deep neural networks are constructed you will be confident that such an eventuality will not be coming without another complete revolution in our approach to the technology. (For just one example I would point to ObjectNet which has managed to fool almost all the previous models that beat human performance on ImageNet).
This is not to say that there are not remarkable advances— there are— they are just not of the singularity type. ML is doing certain tasks far better than humans but is going to need humans to help it do these things, it is not just going to do them on its own.
On what people would consider the more realistic front Andrew McAfee and Eric Brynjolfsson are also wrong, they are too optimistic, machine learning has serious limitations. Some of the best engineers in the world have been working on the problem of creating autonomous vehicles using neural nets for nearly a decade. Tesla is years ahead of all the others and still seems fairly far away from level 5 autonomy (much better than human). McAfee is a bit cringe at a few points, in the book he sounds like one of those hydrogen car people. He talks about cyclopean LiDAR as the most important sensor, as Elon has said LiDAR is just a crutch and he is right, maverick iPhone jailbreaker and founder of comma.ai George Hotz agrees. If you think you need LiDAR for autonomous vehicles you must be an alien because humans drive without lasers all the time. LiDAR is less important than cameras and probably less important than radar. [2]
McAfee and Brynjolfsson writing a little bit earlier in the development of social media were much more optimistic in that area too. In their analyses they use the time people spend on social media platforms to predict the economic value provided to users by said platforms, with the rationale that because the platforms are free it is very difficult to properly assess just how much our lives have been improved by social media in a dollar value. Of course these days it is common knowledge that such websites are attempting feverishly to maintain time on site. Time on site allows the social media giants to show its users advertisements and gain data on their behaviour, they do not care about providing a great service so much as getting you addicted. My argument is that we ought to be analysing who is getting the better deal here — we the people — or the social media platforms. [2]
McAfee and Brynjolfsson are right a lot of the time though, for one innovation is almost everything. They are also right that innovations often come from those you least expect and that there are still plenty of accessible fruit, if we can revolutionise our thinking. Unfortunately they subscribe to recombining view of technology — that massive improvements are made not by brand new ideas but through shuffling old ones around, I will address this soon, it is wrong and right — a kind of half-truth because what actually happens is that there is a break-through and then a lot of recombination that follows but then you hit saturation (see the S-Curve again) and need a new breakthrough or shift in thinking to reveal the new opportunities. [2]
Stephen Wolfram of Wolfram Alpha and Wolfram Mathematica fame describes his views on this subject, “I’ve been fortunate enough to live through some [paradigm shifts] myself, [I’m] even probably responsible for some. I think the main process that seems to happen is a methodological advance is made, that opens up some new area, there is a bunch of low hanging fruit to be picked in that area, lots of exciting stuff happens for 5 years, 10 years, maybe a little bit longer. That area of science gets big, it gets quite institutionalised, it slows down, there is 50 or 100 years of sort of stasis and then maybe there is some kind of methodological advance and it opens up again.” [3]
People underestimated neural networks but that does not mean they are going to keep on going revolutionising everything. If you have tried training a net on your own computer you will understand the computational restraints involved. If the revolution is going to continue and computers are going to conquer tasks even more difficult than autonomous-driving we are going to need a massive breakthrough on the materials science front.
The Moore’s Law Slowdown
People understand exponentiality stfu.
It is now generally accepted that Moore’s law is over. This was inevitable given that the statement “thou shall not be exponential with respect to any physical dimension forever” is basically a law of physics — one it would be nice to overcome but a reality of our present situation nonetheless.
Ironically the source for the graph at left is the futurist Ray Kurzweil’s book The Singularity is Near Clearly he missed the characteristic S shaped curve, growth is slowing.
Intel delayed its 10nm architecture for years and it will likely delay its 7nm for longer. My personal knowledge of semiconductors is fairly weak but there are obvious physical constraints. The fundamental issue is you really cannot go beyond 7nm with silicon because your electrons will start tunneling into adjacent transistors. You may be able to switch from silicon to carbon but even still we are not going to see much beyond another 10 fold increase here and that is going to take a lot of time. [4] [5] [6]
You can increase the number of transistors but my understanding is this causes other issues. Because of the time it takes for information to move around the chip (speed of electric transmission) you cannot simply scale up because this will force you to reduce the clock speed of the chip. Also trying to fit more transistors into a small area will generate heat and that will cause issues if you do not slow down the clock speed. [8]
This is not to say progress has stopped it is just to say progress has slowed down a lot and there is not a lot of hope for making the leaps and bounds we made up until this point within the bounds of our current understanding of physics.
Personal Computing:
Somewhat controversially I would say on the personal computing front the only thing that has really improved the regular user’s experience from a hardware perspective in the last decade is the prevalence of SSDs which dramatically reduce boot times and increased ram which make laptop/tablets far more viable. Not really an awe inspiring technology improvement, (SSDs have improved read and write speeds dramatically but practically for personal computing after you go beyond a couple of gigabits/second, (few hundred megabytes per second), it does not make a noticeable difference). Of course the last decade coincides with far more bloated web-based javascript applications countering some of the benefits of increased ram.
My goal here is not to convince you that information technology is not one of the most amazing if not the most amazing technology in human history — it is just to call into question the default assumption of unbridled optimism moving forward, progress won’t happen on its own, you! the reader! have to drive technological progress by inventing!
What went wrong? Ideology and Determination Pt.1 The Silents and The Boomers
The Silent and Baby Boom Generations experienced a period of unrivaled innovation in their childhoods and early adulthood. The economy was booming, labour was a sellers market and companies could afford to pay. We have all heard stories of the kinds of miraculous things that happened during this period. My Grandfather who worked hard in school got a full scholarship and salary from BHP to attend University and work there just one day per week. When I was younger my father told me the story of him leaving school half way through year 11 and finding a job working for a mechanic almost instantly. This seems foreign today and for good reason.
The BoomerSi Generations took progress for granted and thus became complacent. They did not know what was going to improve or how but they knew it would.
They were as Peter Thiel puts it “indefinite optimists” and the consequence of indefinite optimism is the shift away from engineering to finance.
“[The indefinite optimist] expects to profit from the future but sees no reason to design it concretely. Instead of working for years to build a new product indefinite optimists rearrange already invented ones. Bankers make money by rearranging the capital structures of already existing companies. Lawyers resolve disputes over old things or help other people structure their affairs. And private equity investors and management consultants don’t start new businesses they squeeze extra efficiency from old ones with incessant procedural optimisation. It’s no surprise these fields attract disproportionate numbers of high achieving Ivy League optionality chasers. What could be a more appropriate reward for two decades of resume building than a seemingly elite process oriented career that promises to keep options open…… the strange history of the baby boom produced a generation of indefinite optimists so used to effortless progress that they feel entitled to it. Whether you were born in 1945 or 1950 or 1955 things got better every year for the first 18 years of your life and it had nothing to do with you.” (Thiel, Zero to One, 2–06). [23]
I apologise for ABS’s wonky presentation in this table. As can be seen here the trend toward Society and Culture as well as Business and Administration degrees started much earlier. Whereas Engineering degrees halve from pre-1971 to 91. Even if we assume that the emerging field of computer science was stealing brains from Engineering and the Physical Sciences the total for those three still falls from 16.9% in pre-1971 to 12.4% by 1991.
There is nothing wrong with rearranging structures to make them more efficient but efficiency caps out at 100%, if you want to see real improvement you have to grow the pie not the percentage of the pie you are utilizing.
Which pie do you want? Because there is a trade-off, efficiency constrains and restricts and when you reach 90% plus efficiency and there is very little profit to gain from this method of incremental improvement you will have the big 4 accounting firms paying graduates pennies and trying to distract them from their terrible salary with discounted gym memberships and other gimmicks. As you approach 100% your system will also be highly prone to rent-seeking behaviour under claims of increasing efficiency. There are other more mathematical problems with efficiency we will come to later.
After technological progress began to slow post 1970 income inequality was able to keep the highly successful boomers on an upward trajectory.
Booms, Busts and Bankers
From the period 1920–1970 there were no real crashes outside of 1929 and The Great Depression. There were a couple of blips with Suez and the Kennedy assassination but these were geopolitical events and should be considered differently. The boom bust cycle has only really existed with the current magnitude since the 70s with the first bust being Black Monday, people would like to blame the crash on computers but if it was really just computers then the market would not have taken 2 whole years to recover, the crash revealed that something dysfunctional was going on. Though that being said Black Monday was nothing like the Dot-com Crash and the 2008 global financial crisis. One could have invested in an S&P 500 index when I was born and if they had used a buy and hold strategy would not have made any money until I was 12 years old. If they had instead been long volatility they would have had a great time during that period. There is almost no better time for me to be presenting this argument since Spitznagel’s Black Swan fund has just made a 4000% return thanks to coronavirus. [10] Renaissance Technologies’ Medallion fund performed well during the Dot-Com crash, The GFC and earlier this year during the Rona plunge. [11] Renaissance being famous of course for only hiring mathematicians, physicists and engineers. It is worth asking oneself what has so radically changed since 2012 that justifies the current position of the market, especially considering the situation with the virus.
The Dot-Com crash and the GFC were entirely the making of the financial sector. They are the product of the nouveau científicos (the indefinite optimists). There is a kind of disease in finance that can allow oneself to be deluded into believing in fundamental analysis. If a company has had consistent returns for a long time that can give you very little information depending on the nature of the distribution and the nature of the distribution may not easily be determined from a small number of observations.
Brief Aside on Stats (may want to skip if math makes you feel stupid):
The law of large numbers is a statement about the lim as N -> infinity not about large N (though as a physics student and a fan of Richard Feynman I appreciate the attempt to give it a concrete name). For distributions that have a large kurtosis such as power laws standard methods used on normal distributions are not helpful. Height is an example of an approximately normal distribution. No matter how many people we search for we are probably never going to find someone who is 4m tall — the probability is so close to zero. Therefore if we have a sample of 999 people the 1000th person is unlikely to shift the sample mean much at all. Wealth is an example of a power law [21], if you have 999 people the 1000th person may have a massive impact on the sample mean, if the 1000th person is Bill Gates or Jeff Bezos then the value of the sample mean will increase by many orders of magnitude. In power laws rare events have much more weighting (x * p(x) is much larger because x is much larger). The famous Pareto 80–20 rule is an example of a power-law. Many probabilistic functions in finance have a kurtosis larger than a normal distribution. I was lucky to recently come across a thinker that elucidates these issues clearly his name is Nassim Nicholas Taleb and we will return to him later.
This means an education in finance or economics may actually have adverse affects on someone’s ability to think, reinforcing the idea that the man we discussed at the beginning of the essay, the one telling us that the we needed a degree to understand this more “complicated world”, was a charlatan.
Even after the GFC people still have not learned. There was a group of people called XIV traders so called because they were shorting the VIX (A Volatility Index) this went great for a few years because there was relatively little volatility but that is the point — you cannot predict volatility, when the VIX spiked in 2018 they all lost their money, the market does not suffer fools gladly. [14]
P.S. I do not believe I am at all committing the middleman minority fallacy, the financial sector is rent-seeking and deserving of our ire.
Decentralised Education: Copy what works — The Info Tech Sector
Dropouts
The tech sector — by far the most innovative — is known for not really caring about formal education. In fact Steve Jobs, Larry Ellison, Michael Dell, Bill Gates and Mark Zuckerberg all dropped out of University. Peter Thiel (Paypal Mafia Member and Co-Founder of Palantir) has actively tried to undermine the University Degree system by creating the Thiel Fellowship. Reid Hoffman (PayPal Mafia Member and Founder of LinkedIn) has advocated for a total upending of the College Degree. Elon Musk (Paypal Mafia, Founder Tesla & SpaceX, OpenAI, Neuralink) says he does not care about College degrees only ability.
All the big tech companies are happy to hire those without formal education. They just give them interview problems to see how mathematically capable they are. One amazon interview question is to find the function representing a hanging electricity cable, answering the question in a satisfying way will require you to have some sound intuition and will push the boundaries of first year calculus knowledge (second order single-variable differential equation to solve). Some like the 12 men on a seesaw problem are more of the brain teaser type.
As mentioned previously the most successful hedge funds often look for applicants with no prior knowledge of finance. That is they take an approach similar to the tech companies. Quantitative Trading firms often ask their applicants the kind of brain teaser questions the tech companies often do.
Incentives and Competitions
The tech sector does more than just hire people though it actively encourages innovation and autodidacts through competitions called hackathons. Tech companies big an small hold hackathons with prizes ranging from large chunks of cash to experiences where winners receive tours of their facilities and the ability to talk to the engineers working on the cutting edge of the technology.
Kaggle is a company founded in Melbourne that moved to Silicon Valley and raised significant cash from the likes of Max Levchin (PayPal Mafia) before being bought out by Google. Kaggle has openly accessible datasets for ML and data science autodidacts and holds competitions based on certain datasets and problems. In some of these competitions people who had no previous experience in machine learning outside of watching an online Stanford lecture series were able to produce better models than the best engineers in the industry.
Readers should understand I am not just telling you these things for the fuck of it, maybe you should actually go check some of this shit out.
Hackers: Richard Stallman, Linus Torvalds, Linux and the Open Source Movement
The Open-Source movement is another source of autodidactism encouraged by techies. It is a decentralised approach to programming with the principal that everyone has access to the source code which they are allowed to use (of course you should always credit) for their own projects. Because of this approach it allows rapid progress and avoids transactions costs. There have been many massively successful Open-Source projects.
In the beginning software was essentially Open-Source, people would just pass around code, it was not until around 1980 that software began to become much more proprietary. Of course this ground the gears of many engineers and computer scientists at the time because it meant they could not alter the code for their desired application. The blame for this lies at the feet of some of the big tech firms, Microsoft was one of the companies responsible for the push to proprietary, this is how Gates developed his big bad caricature, there is of course a little bit of merit in this push for proprietary.
Richard Stallman loved the dynamism of the pre-proprietary age of software and was disheartened by how proprietary software was erecting barriers and preventing innovation. So in 1983 he founded the free software foundation. Stallman realised that if he created his own operating system then the decentralised fun could continue. So he created GNU which stands for ‘GNU Not Unix’ a funny way of saying it was like Unix but was not Unix because Unix was proprietary. GNU software is a free copyleft license, here is a description from their website:
“To copyleft a program, we first state that it is copyrighted; then we add distribution terms, which are a legal instrument that gives everyone the rights to use, modify, and redistribute the program’s code, or any program derived from it, but only if the distribution terms are unchanged. Thus, the code and the freedoms become legally inseparable.” -gnu.org/copyleft [15]
“wherever the software goes the freedom follows” — Stallman [16]
“The crucial thing about GNU is that it is free software. Now free software refers not to price but to freedom so think of free speech, not free beer. The freedoms that I am talking about are the freedoms to make changes you want to or hire somebody else to make changes for you if you are using the software for your business, to redistribute copies to share with other people and to make improvements and publish them so that other people can get the benefit of them too. Now those are the freedoms that distinguish free software from non-free software. These are the freedoms that enable people to form a community.” — Stallman [16]
So you have the right to download, redistribute and modify software with a copyleft license but then you too must make your modifications available under the same constraints. The actual license used is a copyleft license known as the GNU public license. If software is proprietary it is possible only one company has access to the source code, so they have a hidden monopoly on support, this is why for example microsoft has such terrible support compared to Linux. This framework created a positive set of incentives for a community of tinkerers to form and allowed what Wolfram described in the earlier quote as roughly ~ “the discipline becomes institutionalised” to be avoided.
Many elements of GNU are still used today. Everyone who has taken an introductory course in C has probably used GCC (GNU Compiler Collection).
Ironically enough very few on the mainstream or even radical left seem to mention Richard Stallman in fact I cannot remember an instance of him being cited by them. Despite Stallman and Chomsky both residing at MIT they did not meet until 2015, they did eventually meet though — something that is worth pointing out. This paragraph is basically just meant to shit on academic leftism and its inability to adopt practical approaches, as the free software movement was the closest thing you would get to successful anarcho-communism in recent history.
Linus Torvalds came to fill in the missing component of the new operating system in 1991. The GNU project had designed the peripheral technologies but the OS (operating system) still needed a kernel. Before the GNU project finished its kernal Torvalds came along with a kernel called Linux and if you are reading this on your phone and it’s an android you are running the Linux kernel right now. Torvalds did not develop the kernel for the GNU project but he had similar ideals about having an open-source approach. Linux then adopted the GNU peripherals and the GPL (GNU Public License) and a new open-source OS was born.
The Linux OS is now the OS used for servers across the world. Linux has won the server market, thanks to Google’s Android OS (which is also open-source) it also dominates the mobile phone market (after the death of the windows phone the only smart phone not running a Linux based OS is the iPhone). The short story here is that Linux is an amazingly powerful technology but it was not created by any bureaucracy it is thanks to a decentralised approach with countless engineers involved.
There were a few hiccups in realising a decentralised approach, Torvalds thought the current services available were holding development back. So he left the Linux project for a few months and came back with Git which was released on the 7th of April 2005. Git is a technology that enables more effective decentralised organisation and code management for large projects.
In 2008 GitHub a code-repository based on Git was founded, on June 4th 2018 is was acquired by Microsoft for $7.5 Billion. Today GitHub or similar platforms are used by almost every programmer and tech company on the planet for their code repositories and projects. A new programmer new to Java such as myself may wish to create Pong as an introductory project, thanks to GitHub I have access to many others who have coded the game and can compare code making it much easier to learn.
Why do we need Universities degrees again?
If you think coding is an exception and most degrees could not be learned through an autodidactic approach you should pause for a minute and understand what you are saying.
Good coding is highly technical and sub-disciplines like machine learning require reasonably complex mathematics relative to other things. Unless your degree is in the natural sciences or engineering there is absolutely no need to do your degree in person at a University. If your argument is about social or soft-skills for commerce these could be developed from experience in industry. As for the humanities and social sciences it is obvious it would be far easier to teadch them online than to teach coding online. As for the consumption good function of the University that could easily be served in a different setting, young people will find ways and reasons to party. If your argument is that you require the University to impose deadlines on you and it provides an incentive structure then we should be asking why we need to saddle people with the threat of tens of thousands of dollars of debt just to get them to complete their degrees and maybe if you need to be threatened like that you should not be taking the course in the first place.
Well maybe the University makes you a good researcher? I would be highly sceptical of this seeing as for most undergraduate subjects with scope for research essays (arts) have predetermined readings and essay questions are mandated. Well maybe University… you see there is nothing really the University does for students they could not get from a variety of separate services that would do those jobs much better and for much less.
Democratisation of information => Monopolisation of attention
The natural consequence of exposure to the internet will be the obliteration of 90% of Universities (provided market forces are permitted to prevail of course). Take a look at sports or music, after they were exposed to mediums that allowed large numbers of people to view them at their convenience, television and the internet respectively, a very small number of athletes and performers captured most of the attention. People would rather watch the best of the best, than have an excessive abundance of just good, it makes sense after all people have a finite amount of time. The same trend prevails in Hollywood actors and unfortunately increasingly with CEOs and Politicians.
Provided regulation does not get in the way then this too will extend itself to University degrees because there are massive efficiency gains. Why would you learn from some second rate lecturer who would prefer to be in the lab or doing research when you could learn your linguistics from Ivy League Professors like Noam Chomsky or John McWhorter (host of lexicon valley podcast, check it out), who are both great at articulating the core concepts and are at the forefront of the science. Imagine and I think it is possible — that someone was able to create a faithful recreation of all the Feynman lectures that would make redundant all the first and second year physics lecturers and clearly too the benefit of the students.
In a world of free markets the undergraduate business will depart the University and that is the healthiest outcome in my view. We would prevent any further students in the United States from getting wrapped up in the Student Debt Crisis which far from increasing social mobility is likely to perpetuate inequalities. It is too obvious why has it not already happened?! Good question do not fret we shall return to answer this question but I hope at least the reader is beginning to see a theme emerge.
The Last Step: Coming After Physics — 50 years of not so much?
Disclaimer: A lot of what I am about to say does not apply to the Astrophysicists and the Condensed Matter people.
What is a physicist?
A physicist is someone who looks to see if current theory is closed over itself and reality under some sensible operation. This function cannot be well performed by a priesthood.
Case 1:
Isaac Newton was in every sense an outsider at Cambridge. Most of his greatest work was done when he was hiding away in his room at Trinity being an utter recluse or during the plague years when he returned home. Newton worked outside the establishment — he had to, he had a definite plan to tinker and find out the laws of the Universe. His work on optics and his later work as an investigator of monetary fraud proved that he had a very practical hands on mind, his head was not all in the clouds.
Case 2:
James Watt was an outsider too. He earned his prestige through his inventions, most important of which was the massive improvement he made to the efficiency of the steam engine. Watt never received a University degree, he did an apprenticeship as a scientific instrument maker in London for about a year and then returned to Glasgow to start up a shop selling instruments to scientists. He distrusted publishing papers and preferred to pursue patents instead.
Case 3:
Benjamin Franklin was another outsider. Franklin received the Copley Medal from the Royal Society and honourary degrees from Harvard and Yale but he never attended University. His inventions stretch far and wide and obviously he made many contributions to politics and was a great statesman, it is worth saying this was by no means evident from birth — Franklin was social climber of the highest order, the original American Dream.
Case 4:
Michael Faraday too was an outsider. He came from nothing, he learned what he knew about the world from his job working in a printing house where he would read the sheets of paper as he assembled them into books. He managed to get a job at the Royal Society by impressing Humphry Davy. Einstein kept a picture of him on his wall, alongside Newton and Maxwell. Faraday is notorious for lacking any mathematical skill whatever, but as Rutherford put it,
When we consider the magnitude and extent of his discoveries and their influence on the progress of science and of industry, there is no honour too great to pay to the memory of Faraday, one of the greatest scientific discoverers of all time.-Rutherford
Faraday was the greatest tinkerer of all time. Among many other great inventions and ideas the electric motor stands out as it revolutionised not only our society but our understanding of the universe (remember this man had no formal education).
Case 5:
Nikola Tesla was an outsider. Tesla dropped out before finishing his degree. He built on the work of Faraday and constructed the first alternating current motor, generator and transmission system. If you are reading this online you are on a Tesla powered device, if you have a battery powered device it was charged by a Tesla grid. The work of Faraday and Tesla — outsiders — kicked off an engineering revolution that powered the growth from 1920–1970.
Case 6:
Einstein was exceptional and like these others he was not a big fan of University — he passed. In fact later in life he would leap at the opportunity to help set up the Institute for Advanced Studies where guess what — there was no teaching and they accepted no one with an undergrad degree! Some of his best work was done when he was at the patent office and there is a famous quote from his time there.
“The origin of all technical achievements is the divine curiosity and the play instinct of the working and thinking researcher, as well as the constructive fantasy of the technical inventor.” -AE from time at patent office.
“Scientists investigate that which already is; engineers create that which has never been.” — Albert Einstein
Stopping here is arbitrary I could continue but I don’t want to bore you to death.
SuperString Theory A Whole Lotta Nothin?
In the last 50 years physics has become institutionalised. This coincides with very few engineers working on problems relating to theoretical physics. Supersymmetric particles were not observed at the right energy levels in the large hadron collider. How did we get here?
In the last 50 years in theoretical physics there has been an entirely self-policing community. This lead to expert problems, peer-review does not help if you are not connected to the outside world in some way. In fact peer-review is a scam, it is a recent addition and is used to suppress new ideas, the right ideas will rise to the top, physics does not really need peer review.
Ed Witten is the super genius that turned theoretical physicists into mathematicians. His merging of the various Superstring theories to many people seemed too good to be true and so a generation of progress was loss down this particular rabbit hole, there are some remarkable things about this theory — for example if it were correct it predicted a WIMP (weakly interacting massive particle) that would have accounted for dark matter.
If I can stretch my argument to the Queen of all sciences being distorted and warped by the modern University then how much more do you want from me?! Huh? Do not worry the synthesis is coming soon.
Ideology and Determination Pt 2, The Millennials — Definite Pessimism
Most economic fallacies derive from the tendency to assume that there is a fixed pie, that one party can gain only at the expense of another.- Milton Freedman
The Millenials are definite pessimists they know the world is going to get worse and they know how it is going to get worse. Climate Change, Global Capitalist Exploitation, Big Corporations, Alienated Labour, Automation and Wealth Inequality. Lacking any creativity they see the government as the solution to everything. They like their Boomer parents believe in societal institutions but only the ones they agree with, the leftist media, the Universities, HR Departments. They hate their predicament so much that they have turned against society, with their postmodern philosophy they seek to seize control of institutions to get what they want, not what is good.
I will refrain from going to deep on this subject because it deserves its own 10,000 word essay just to begin to list the fallacies and contradictions present in the ideology of the generation. For now I will suffice to show that it is impossible for Governments, Universities or Corporations to give them what they want.
In summary humanity’s current course is unsustainable — this is what the millennials get right — but what they do not get is that you will never convince people to lower their standard of living and if everyone in the world wants the standard of living we have in the West with current technologies no amount of government intervention is going to make that work the only way out is innovation. You can prove this purely from the perspective of resources — and your paper/metal/no straws are not going to fix this.
Taleb — Antifragility: Bigger is Badder and Smaller is Sexier
Damocles, The Phoenix and The Hydra:
Antifragile is a word that has been introduced into the English language by Nassim Nicholas Taleb, as he could not find a word for the kind of phenomenon he was trying to describe, since there was no word for the kind of thing he was trying to describe in English or in several other languages. There is some intuition building required for this topic so I will introduce it in a similar way to how Taleb does in his book, with my own spin on it. [22]
Damocles
The story of Damocles comes down to us from Cicero in his Tusculanae Disputationes, the story takes place Sicily when it was under the rule of Dionysus II who Plato had a go at taming and training into a Philosopher King, he got himself imprisoned for his trouble.
Damocles got himself in trouble with Dionysus II by praising him and how happy he was a little too vigorously. Dionysus II was massively paranoid and so did not live a very happy life. To help Damocles realise how unhappy he was he had Damocles lie on a bed of gold embroided and wrought with high taste, he then ordered some good looking youths to wait on a table and give Damocles whatever he desired, etc etc. Damocles basically had all the luxuries he could possibly desire. Damocles thought this was pretty good, but into this arrangement Dionysus introduced a sword to hang down over Damocles head from above suspended by but a horse hair. The end of the story is interesting but it diverges from our interest in how we can use it as an analogy here. Damocles in this state has all he could desire and as long as he can forget about the sword he has a very stable and pleasant existence provided of course the hair does not snap. This is the incomplete analogy for fragility but serves to help build the intuition somewhat, Damocles life is highly subject to disorder.
The Analogy for stability is the Phoenix. When a Phoenix dies it bursts into flames and it is reborn from the ashes. Whatever you do you remain static to disorder.
Finally the analogy for antifragility is the hydra. If you cut of the head of a hydra, then two more grow back. The hydra is not simply robust, resilient or stable it is antifragile. The Hydra gains from disorder, the more disorder the hydra is exposed to the stronger it becomes. We want the human society to be a hydra.
Examples of real life antifragility:
Mithridates of Pontus is a hell of a character I recommend looking up. Anywho for reasons we won’t go into here the man was one of the first to discover you could build up a tolerance to poison. Over a period of many years he slowly increased the dose such that if someone tried to poison him he would be able to survive. In this example the body is the antifragile sytem as it overcompensates — it does not simply resist the poison it overcompensates becoming capable of enduring larger doses.
Another example is weightlifting, after a session of training the body overcompensates such that next time you lift you can lift more than you could lift before.
Fasting is also an example where we see antifragile effects. There are many documented benefits of fasting and caloric restriction, so that afterward the body is actually healthier than prior to the fast.
Temperature variation — saunas and cold showers — also seem to have beneficial effects.
Many biological systems display the characteristics of antifragility. Antifragility is an essential element for any self-replicator if it is going to survive in a Universe where entropy is always driving things toward disorder. Stability is not good enough, if you aim for stability in a Universe where disorder prevails you will fail, but if you aim for anti-fragility you may get stability or even better.
The innards of the beasts
How do you create an antifragile system?
Let us look at the Elephant and the Mouse. Kleiber’s Law states that metabolic rate obeys a power law. If q is the metabolic rate then q ~ M^(3/4). This means it is better to bigger if you take a naive approach as you will be more efficient per unit mass (sublinear scaling). It turns out the circulatory, neural and respiratory systems also follow a quarter power law.
Geoffrey West Particle Physicist turned Biologist (see also Enquist and Brown) argues this is due to network existing in 3 dimensions and trying to fill the space leading to a fractal geometry and in a sense an attempt to stretch out into the 4th dimension leading to integer multiples of power 1/4.
If you want to understand the idea here I recommend this 3Blue1Brown video on space-filling curves: https://www.youtube.com/watch?v=RU0wScIj36o
If you want to go even a little deeper than that you should check out: https://www.youtube.com/watch?v=gB9n2gHsHN4
West states several key principles that unite the biological systems he studies: All biological systems function by way of networks, which transport energy, matter or information.
1: The network is space-filling (fractal geometry)
2: Terminal Units Are Invariant (Capillaries in a blue whale ~ in size to capillaries in humans, for example, analogy is imagine if electrical plugs scaled with building size — it would not work, capillaries are servicing cells of similar size therefore they too must be of similar size)
3: That over time biological organisms become optimized over the course of evolution
[17] [18] [19] [20]
Since biological systems tend to be antifragile with respect to many kinds of disorder we should seek to replicate these principles in our society at large but before we move onto humans we should say a little more about elephants.
Mice vs Elephants
If a meteor were to hit planet Earth mice have a decent chance of surviving whereas elephants do not. In fact it is believed that mammals survived the meteor that killed the dinosaurs because a small shrew like creature was able to survive. It is suspected this is because the meteor induced a nuclear winter and there was not sufficient food for megafauna to survive, it is worth noting in the medium term megafauna do perform better though because they can fast for extended duration but even though they are metabolically more efficient they still require a larger baseline energy consumption and so if resources are made more scarce then their populations will suffer or go extinct. There are also many more mice than elephants which trivially makes it more difficult to kill them all. Another advantage for mice is that they are immune to fall damage, you can throw a mouse out of a window and they will be fine this is because of the surface area to volume ratio, mice have much more surface area to distribute the mass of their insides over than elephants do, this will mean that elephants will explode if you drop them from high enough and mice will be fine. The surface area to mass ratio has consequences for terminal velocity meaning that elephants hit the ground much harder as well.
While it is true that individual elephants are much more resilient to many forms of disorder in the short term than mice are, they are much more susceptible to fat-tailed risks or extreme events like climate change. Most of the major mega fauna died out at the end of the ice age, it is suspected that humans were partly responsible but for many it was because their baseline energy consumption was too high when the climate warmed many had an insufficient energy supply and went extinct, so their increased metabolic efficiency did not matter. So to have an antifragile population actually means distributing the risk of the population out to the individuals.
Of Elephants and Men
To put it simply in the short-term there is a trade-off to being antifragile, you will lose efficiency. The nature of antifragility means that long term you will actually make more progress because your system will gain from disorder whereas the fragile efficient system will collapse. To overcome the disorder you will need a radical restructuring which will likely kick off an S-Curve growth spurt. This means if you become institutionalised you can be very good at what you do right now, in fact better than if you continued to innovate but after exposed to a certain form of disorder things will begin to break down, that can happen quickly through something catastrophic or slowly through decay and corruption.
If you look at our society at present we are highly geared toward efficient fragile systems, for example constant M&As to increase efficiency at the cost of long term risk to society at large — because the entire industry could be lost, which of course cannot be stomached, we then have to bailout the corporation — thwarting the market which is telling us that the company/companies in question are too large.
Keynesianism and the deployment of MMT is a disaster — the economy is in the worst state it has ever been in because we decided to bail out corporations that were fragile and which should have gone bust, (the banks and the big auto makers).
Keynes lost the debate with Hayek by mistaking the symptom for the disease, a sudden lack of liquidity in the market does not just happen it is caused by something — in the case of the GFC it was caused by the financial sector by using quantitative easing in this scenario you are perpetuating a moral hazard. If we keep medicating the economy to suppress the symptoms while the disease continues to spread then eventually the dollar will collapse under the weight of its own delusion — I will leave this comment here because I am not sufficiently educated on macroeconomics but this seems so obvious I cannot imagine any theory that would convince me otherwise.
We also see fragility in Just-In-Time manufacturing and supplying, if one element of the supply chain has a mishap entire production ceases until that issue is resolved. Half the things in this essay I had not even heard of before I just predicted it from the theory of antifragility — it is obvious that it should not work — google it and sure enough people are complaining about how Just-In-Time production broke down during the fat-tailed catastrophe of 2011 the Japanese Earthquake and Tsunami. The generality and simplicity of the theory is what makes antifragility so compelling.
The implication of the trade-off between antifragility and short-term stability is little innovation and highly standardised and regular outcomes — I feel like I have to keep saying this in the short term!. This is what I have sought to illustrate with all the previous examples in this essay. Other industries that get subsidies or tax rebates for dubious reasons will have the same outcomes. Universities are one such institution that is allowed to exist despite it being incredibly fragile and deleterious to society. Universities now train people to become incompetent and incapable of dealing with the real world while dialing up their hubris. They are the organs of highly standardised outcomes — they are unlikely to produce revolutions in any technology if they do it is because the researcher succeeded despite them. The tax break is like being constantly bailed out relative to the rest of society and through regulatory frameworks (see TEQSA) they maintain their grips on degrees and thus accreditation, Universities are socialism!
We know big is bad, that distributed networks display antifragility and centralised ones do not, simply because centralised systems centralise fat-tail risk.
Finding a fix:
Fractal localism — The Political Solution — The 5th Attractor (Game B)
In fractal localism, localities have power over their surroundings, states have power over issues that collectively concern their localities and countries have power over issues that collectively concern the states, 3 levels is arbitrary it is just that we have names for those 3 levels it could be 7 levels or 9 levels there will be diminishing returns though. If you want another example of bad governance because a nation level governing body thinks it knows everything look up NHS vs Scottish NHS (hint: they devolved the power to a lower level and it functions better).
I could keep rattling on about this forever but I feel I have given sufficient examples of how things go wrong when you have large short term efficient fragile systems and how things can go amazingly well when you do not.
So I will basically leave it at this, there are no positive rights at the federal level that is exceedingly stupid if there are they can certainly not be administered by a central authority, at best they can be overseen, to advocate for central authority is to enshrine fragility into the system from the get go. The further away you are from the terminal unit the more broad and inconsequential your policies should be for that terminal unit (whether it be the family or the individual).
Government needs to get out of business too. No more bailouts, no tax breaks (one uniform logical tax system), no subsidies for any industry. It needs long term strategic planning an financing for very specific things. The federal level is for the military and foreign policy, the federal level should have very little to do with domestic policy outside of large infrastructure projects.
On the side of corporations a lot of thought needs to go into rethinking the structure of the current market and publicly traded assets. People get crazy about changes, I do too because normally they are interventions that distort natural pressures but some of the systems in the economy we have accepted for default and without really asking the deep question of whether they serve the axioms of a free markets, systems that do not need to be modified.
Ideology and Determination Pt 3 The Greatest Generation and the Zoomers or Bloomers?
I hate people who complain all the time without presenting solutions. So this will be the solution to the ideology question.
Very few people read past the first few pages of a book so you should always put the juicy stuff at the start. King James Genesis begins with God creating the world from undifferentiated nothing with the spoken word he willed it into existence and it was good. God created man in his image — and all this in the first chapter of genesis. In the time of the Greatest Generation the church was a much larger part of people’s life there was a belief that creation itself is good and noble. The universe is not morally relative not all combinations are morally equivalent and moving the universe towards some combinations is good. If one looks at the world this way one can be a definite optimist, God had a plan for creation and he executed it and it was good and he made humans in his image, which means that humans can create things that are good, (one should be careful not to go too far with this and actually become religious).
The second thing that characterised the Greatest Generation was that they believed in their countries and their local communities. They had skin in the game. Their parents had lived through WW1, some of them did not know their fathers because they had died in the war. Many of them would later fight in WW2 themselves. They had shared experience of tragedy and were bound together by it. Politicians also had skin in the game, their children would go off to fight in the wars, Churchill wanted to have his skin in the game and did at several points. He planned on watching the D-day landings from HMS Belfast but King George VI persuaded him not to by threatening that he would go with Churchill.
This is another idea of Taleb’s in fact he a whole book named after it. If we want to get back in the business of innovating and become prosperous once more then we need to create structures where people have Skin in the Game. Thiel’s advice for venture capitalists and entrepreneurs is that the CEO should always earn the least, that way he has the most skin in the game — the employees will know he believes in the success of the enterprise as his entire pay off will be from equity when the company succeeds.
It was not until the second half of the 20th century that the protestant work ethic was torn down and America stopped being a single country and became two. The puritan belief in hard work and creating new things is what we need but all things in moderation. We should all try to be a little bit more Ben Franklin.
P.S. I never got to mention that there were more startups in the US in the 1980s than there are today despite massive population increase.
References:
[1]: C.(2017), Australians pursuing higher education in record numbers. [Media Release]. Australia Bureau of Statistics.
https://www.abs.gov.au/AUSSTATS/abs@.nsf/mediareleasesbyReleaseDate/1533FE5A8541D66CCA2581BF00362D1D
[2]: Brynjolfsson, E., McAfee, A. (2014). The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. United States: W. W. Norton.
[3]: Dr Bryan Keating’s Youtube Channel (2020) Stephen Wolfram & Eric Weinstein: The Mathematical Nature of Reality, https://www.youtube.com/watch?v=OI0AZ4Y4Ip4
[4]: Gartenberg, C., 2020. Intel’S Next-Gen 7Nm Chips Are Delayed Until At Least 2022. [online] The Verge. Available at: <https://www.theverge.com/circuitbreaker/2020/7/23/21336356/intel-7nm-chips-next-gen-delay-q2-2020-earnings-amd-10nm-tiger-lake-desktop-laptop-cpu> [Accessed 29 August 2020].
[5]: Gartenberg, C., 2018. Intel’S 10Nm Cannon Lake Chips Are Delayed Again. [online] The Verge. Available at: <https://www.theverge.com/circuitbreaker/2018/4/27/17291040/intel-10nm-cannon-lake-chips-delayed-2019-cpu-processor> [Accessed 29 August 2020].
[6]: Gartenberg, C., 2020. The World’s Smallest Transistor Is 1Nm Long, Physics Be Damned. [online] The Verge. Available at: <https://www.theverge.com/circuitbreaker/2016/10/6/13187820/one-nanometer-transistor-berkeley-lab-moores-law> [Accessed 29 August 2020].
[7]: Kurzweil, R. (2005). The Singularity Is Near: When Humans Transcend Biology. United States: Penguin Publishing Group.
[8]: Fox, A., 2018. Why CPU Clock Speed Isn’t Increasing — Make Tech Easier. [online] Make Tech Easier. Available at: <https://www.maketecheasier.com/why-cpu-clock-speed-isnt-increasing/> [Accessed 29 August 2020].
[9]: C. (2006), Educational Attainment: People with degrees. [Media Release]. Australia Bureau of Statistics.
https://www.abs.gov.au/ausstats/abs@.nsf/2f762f95845417aeca25706c00834efa/b061f670b1b80565ca2570ec00786345!OpenDocument
[10]: Gara, A., 2020. How A Goat Farmer Built A Doomsday Machine That Just Booked A 4,144% Return. [online] Forbes. Available at: <https://www.forbes.com/sites/antoinegara/2020/04/13/how-a-goat-farmer-built-a-doomsday-machine-that-just-booked-a-4144-return/#53b9f60a3b1b> [Accessed 29 August 2020].
[11]: Dewey, R. and Moallemi, C., 2019. Bloomberg — Are You A Robot?. [online] Bloomberg.com. Available at: <https://www.bloomberg.com/news/articles/2019-11-12/the-unsolved-mystery-of-the-medallion-fund-s-success> [Accessed 29 August 2020].
[12]: Taleb, N. N. (2020). Statistical Consequences of Fat Tails: Real World Preasymptotics, Epistemology, and Applications. United States: NASSIM TALEB. pp26
[13]: Taleb, N. N. (2020). Statistical Consequences of Fat Tails: Real World Preasymptotics, Epistemology, and Applications. United States: NASSIM TALEB. pp27
[14]: Langlois, S., 2020. XIV Trader: ‘I’Ve Lost $4 Million, 3 Years Of Work And Other People’S Money’. [online] MarketWatch. Available at: <https://www.marketwatch.com/story/xiv-trader-ive-lost-4-million-3-years-of-work-and-other-peoples-money-2018-02-06> [Accessed 29 August 2020].
[15]: gnu.org/copyleft
[16]: Revolution OS Documentary 2001
[17]: West, G. B., Brown, J. H., & Enquist, B. J. (1997). A general model for the origin of allometric scaling laws in biology. Science, 276(5309), 122–126.
[18]: West, G. B., Brown, J. H., & Enquist, B. J. (1999). The fourth dimension of life: fractal geometry and allometric scaling of organisms. science, 284(5420), 1677–1679.
[19]: West, G. B., Brown, J. H., & Enquist, B. J. (1999). A general model for the structure and allometry of plant vascular systems. Nature, 400(6745), 664–667.
[20]: Bettencourt, L. M., Lobo, J., Helbing, D., Kühnert, C., & West, G. B. (2007). Growth, innovation, scaling, and the pace of life in cities. Proceedings of the national academy of sciences, 104(17), 7301–7306.
[21]: Levy, M., & Solomon, S. (1997). New evidence for the power-law distribution of wealth. Physica A: Statistical Mechanics and its Applications, 242(1–2), 90–94.
[22]: Taleb, N. N. (2012). Antifragile: Things that Gain from Disorder. United Kingdom: Penguin Books Limited.
[23]: Masters, B., Thiel, P. (2014). Zero to One: Notes on Start Ups, Or How to Build the Future. United Kingdom: Ebury Publishing.
