Gill Eapen's Blog, page 78

Learning to fly

Flying for humans have always been about rockets, jets and the Bernoulli effect. This has worked well and this has managed to shuttle people around the world, albeit spewing an incredible amount of greenhouse gases. It is, however, an unnatural way to fly – something that may have been obvious to those who first observed flying birds. The recent paper by Mr. William Thielicke in the Society for Experimental Biology Annual Conference in Glasgow, illustrates why we may have to learn to fly again.

Flying, as most recognize, is about take off, coasting and landing. Technologies that disproportionately help one aspect or the other is unlikely to be optimal for the whole flight. Wing flapping propulsion and highly maneuverable guidance systems may fundamentally change flying, possibly merging all transportation systems – both short and long range. It may also allow a higher level of customizability, moving us closer to anytime, anywhere personal transportation systems. The advantage of fixed wing designs has always been said to be efficiency at the cost of maneuverability. However, these comparisons are made across available technologies in various design spaces. Thus, they are not always true although it may have held many aerospace engineers prisoners to what they have been taught in closed classrooms. They may also been influenced by engineering behemoths who have significant incentive to perpetuate the status-quo.

It is time engineers attempted to learn to fly again. There is no single way and what is available now is unnatural and artificially "efficient," if societal costs are not fully internalized.

Loaded for bear

Contributed by Dr. Ian Williams

"Earlier this month, bear hunters in Oregon stumbled on what turned out to be the largest marijuana plantation that state has ever seized, by far - more than 91,000 marijuana plants growing in a remote canyon in the Wallowa-Whitman National Forest, that's in the northeastern corner of the state. Six armed growers were arrested. Police removed or destroyed the pot plants."

This was the lead in to a story reported by NPR earlier this week. The reporter interviewed the wildlife officer responsible for the clean up. The tone used by both participants in the interview was as if a stock pile of depleted uranium had been found. Environmental insults such as fertilizer and compacted ground were spoken of in language typically reserved for mass murder or genocide. The park ranger reserved his most strident condemnation for the fact that his team would have to remove the sleeping bags and cooking utensils used by the pot growers 

Two important points were never discussed. Firstly, this situation resulted solely because of  the US's medieval criminalization of marijuana, a drug which is much less harmful than tobacco or alcohol and legal in many developed countries. Secondly, no mention was made of the environmental havoc wrought by people tramping into the wilderness to shoot bears. The damage to the ecosystem caused by killing these large herbivores is enormous and yet the story suggested that these hunters were society's saviors for uncovering this festering den of pollution and corruption.

The US could save millions of dollars instantly by legalizing cannabis. It could also deprive organized crime of an important source of income and free up many customs and law enforcement officers to tackle violent crime.

Oh well, in the meantime I'll savor my scotch and cigar while I keep my gun loaded for bear.

Retireless

The concept of retirement is relatively recent and it appears to be a proxy for "escape," from the trials and tribulations of modern life style. Most societies are now organized to aid this – work hard till some point and then drop out of view (retire). Both from the individual and the societal perspective, this is a problematic concept for a variety of reasons. For starters, the stress of upcoming retirement is becoming too much for most of the population. Such stress leads to disease and eventually less happy retirement. Retirement should really be an option for the retiree and not something she has to do. Retirement planning is big business and a large part of the economy is organized around this, although there is little evidence that it adds much value and so many have a reason to perpetuate the idea.



More importantly, the idea of leading a stressful life till retirement in anticipation of a happier life simply does not make sense. If one is happier while working it is likely that she will continue without any consideration of retirement. The desire to retire, then, has to be a symptom of undesirable pre-retirement life and jobs. From a planning and policy perspective, it is important to understand why people want to retire and not just how to pay for the retirees. Many can add significant value to themselves and to society if they forget to retire. There is also empirical evidence that people with active working lives are less prone to disease.



One way to solve the impending "retirement problem," is to eliminate the idea of retirement. Increasing the retirement age has been suggested as a solution but this just prolongs the agony and makes the problem worse. Policy makers have to ask why people want to retire and understand the causes of retirement to aid the development of better designs.

Storage options

Much focus has been put on alternative energy production methods and not enough on electricity storage. Electricity remains to be one of the few commodities that does not allow efficient storage in large quantities. As the universe of production methods shift to renewables – many of them with intermittent production capacity, storage becomes even more important for the functioning of the overall system. The option value of storage increases dramatically as the marginal cost of intermittent production sources trends lower, such as solar and wind. Energy will be produced by a solar plant when the sun shines regardless of the immediate need for it. If it cannot be used immediately, in most cases, it will go waste or even create havoc on the grid, if no storage is available.

Storage, thus, is at least as important a consideration as the production method. Technology does not seem to be advancing fast enough in this area. The focus cannot be just on centralized storage as distributed storage is an equally viable option. In this context, transportation systems that uses energy could use their storage mechanisms as distributed storage if the hardware is available. However, there is a danger here as this may arrest improvements in battery technologies as many will be tempted to put current batteries into action. Chemistry based batteries have always been inelegant and inefficient and these thus far have taken the focus away from capacitors and magnets as possible sources of storage. This is yet another example of investments seeking incremental improvements in conventional technologies while completely discarding possibilities that are an order of magnitude better.

Storage of energy is at least as important as producing it. It is falling behind both due to a lack of focus as well as biases that favor chemistry based technologies.

HDL out, LDL in?

A recent large study demonstrates that the conventional wisdom of increasing HDL and lowering triglycerides do not help arrest the progression of heart disease. This is consistent with the general observations from the applications of many classes of synthetic chemicals – robust conclusions drawn from controlled and smaller populations simply do not hold when measured in large and uncontrolled populations. It also shows that killing one organ (say, the liver) to protect another (say, the heart) is not necessarily a good strategy. This may have been obvious to many ancient populations – but it appears to be the modus operandi for modern medicine.

There are two major problems in the way medicines are discovered and applied today. First, health has to be considered at a systemic level and not at a component and disease level. This is the legacy that every contemporary technical profession – from medicine to engineering-  has left for the future generations. Everybody has been trained to "break down," the problem into its components and solve them – element by element, issue by issue and symptom by symptom. This "scientific method," that we are so proud of is wrong and has led us to bad solutions in every field from high energy Physics to economic policy, in addition to medicine.

Second, statistical analysis and hypothesis testing – the "tools of discovery," in the modern world are fraught with problems. These are tools of convenience – immense help for those wanting to "discover something," as fast and as efficiently as possible and push scientific papers out as quickly as the publishers can accept them. Since these tools have become mainstream, human creativity and the rate of discovery have just taken off. Suddenly the generation with "statistical tools" are so much more adept at "discovering," things. This is not so – we are not any more adept at discovering things – we are just so much better at proving that we have discovered something. In medicine, modern therapies have substantially failed to increase health and utility for humans.

Unless scientists are willing to learn from the "past," and humbly recalibrate – throwing out the fake tools handed out to them and challenging the wisdom of their own coveted process," true rate of discovery will continue to decline, shuttling us back to the dark ages, if we are not there already.

Slower Artificial Intelligence

Most involved in artificial intelligence research and practice would like faster computers and software. The basic premise remains to be that the brain is a very fast and efficient computing machine and to replicate intelligence will require significant improvement in speed and size of computing. This makes sense. However, such a process is a linear extension of traditional computing ideas in an effort to explain complexity, that is not well understood.

It is conceivable that the brain is not a fast computing machine at all. However, it demonstrates the ability to assimilate complex information quickly to produce outcomes that are extremely impressive. There are two assumptions in the previous statement – first, the brain is assimilating complex information and second it is producing outcomes that are impressive. The first assumption is an observation that the brain has many channels through which it appears to be collecting information but it is not exactly clear if it is using all that information. An alternative hypothesis may be that brain is using only a small percentage of the available information. If the discarding of information is systematic, it will require high computing power but it is also possible that the brain is using simple rules of thump and indiscriminately discarding information because it simply does not have the computing power to process it.

The second assumption that the outcomes the brain produces are impressive is debatable– the human brain perceives its own process as impressive but that is not an absolute. For example, if the brain is selecting from available behavior and outcomes templates (that are limited) based on the inputs, the fact that the templates are complex does not mean that it requires computing intensity. Artificial Intelligence focuses on creating the templates from scratch based on a large amount of input information but the human brain may not do that at all. The operating system, at birth, is loaded with some basic templates and over time the brain makes some changes to these templates. But the outcome itself is really a selection problem (i.e. which template to produce) rather than a design problem (how to design the template).

Many have been fascinated by the workings of the human brain and many have been toiling for the past few decades to try to replicate it using traditional computing – by making the machines faster and their memory larger. It may be worthwhile to step back and challenge the basic assumptions underlying this effort. It may be possible to reproduce the outcomes of the brain by simple processes that require low computing intensity and memory. If so, such intelligence may not be worth replicating.

Child options

The authors of some of the recent work on "happiness," appear to be perplexed by certain human actions, such as having children, that do not appear to increase happiness. Based on this, they hypothesize that humans are driven not to maximize happiness but other unknown and mysterious needs. This does not follow. This is the same problem that decision-makers in companies face when they forecast cash flows and discount them to calculate a Net Present Value (NPV). Often such NPV is counter-intuitive and many make up "unknown and mysterious" strategic value to justify their decisions. The reason is that, NPV or Net Present Happiness for that matter, cannot be calculated by discounting forecasted future streams.

For example, having children, give parents options – ability to make decisions as the future unfolds. In the agrarian societies of yesteryear, having children was akin to what corporate finance professionals currently do – children are like cash flows that can be discounted to the present as their services will be deployed in the field and the benefits of such services show low uncertainty. This is not the case anymore – benefits of children show high volatility but having children also provide parents flexibility. Children, thus, represent basket of options held by the parents – the value (in terms of happiness) can only be determined by options based analyses and not by deterministic calculations.

The perils of deterministic methodologies seem to be spreading to other areas as well – including psychology, policy and societal designs - in addition to corporate finance.

Exponential ignorance

Contemporary futurists in virtually every field are fond of invoking "exponential growth," in imagining the world of tomorrow. Some see computer smartness increasing exponentially and Watson's babies making humans distinctly inferior. Some others see medical technologies able to diagnose, treat and re-treat human diseases before they occur, aided by the information explosion from the human genome. Yet others envision transportation systems that shuttle people across interstellar space. All of these visions of the future are aided by a simple premise – an "exponential increase" in information and technology. A few have even banded together to form the "singularity university," to teach people how such exponential increase in knowledge will transform everything in a few years, if not in a few days.

Information and technology have been increasing "exponentially," for a while. Although every generation imagined the future in a similar vein, nothing much has happened. Humans are extremely predictable animals – they will create and destroy, think and sleep, nourish and mutilate and band together in cults and countries to throw stones across the borders. "Exponential increase," in kn0w-how does not affect anything much – it certainly did not improve aggregate societal utility in the last century and it is unlikely to do so in the next, just because futurists so desperately want it so.

For example, the "exponential increase" in medical technology, thus far, has extended human life span in certain quarters and not so in others. Where it extended life span, it did so sub-optimally handing out additional time covered in pain and tribulation. Exponential or not, such improvements, although intellectually stimulating for some, have little value for society. Humans get older and weaker, regardless of all the "exponential increase in knowledge, " till they eventually give up and wither away – this has happened for 50,000 years and it will likely happen for another 50,000 even if futurists come up with robotic pills that can take pictures of the intestine and introduce nanotechnology based micro doses of chemicals so precisely to every nook and corner of the human body.

When the first human in the African savannah looked up in the sky and witnessed the shooting star, she might have gotten the idea that burning something and attaching it to a much smaller payload will propel it across the heavens. 50,000 years later, the "exponential increase" in space know-how got humans to burn many tones of fossil fuel to project a peanut into space. A few thousand years from now, with exponential technology, they will be doing the same – perhaps a bigger peanut with less burning - but fundamentally the same.

Artificial intelligence has been in the air for nearly 30 years now – some dreaming up intelligent robots and others fearing a complete take over by machines. "Exponential increase in machine intelligence," has not yet created something that is as stupid as the humans – something that is irrational and timid, something that will kill its neighbor for tactical monetary gains or something that will organize around religion or such notions. Machines remain vastly superior in-spite of the feeble attempt to change them into humans.

Humans are more likely to reach singularity by stupidity and not by exponential increase in knowledge.

System reboot

 

Based on the purchasing power parity (PPP) corrected GDP, the US, China and India are expected to dominate the world in the coming decades. This also means that these three countries have significant responsibilities to help create a sustainable environment for the world as well as shape the implementation of economic systems that are not biased and provide equal opportunities for everybody. Although many have been critical of the US, it is clear that it has been carrying a disproportionate share of this responsibility.

Let's look at the economic systems for example. Although there have been many missteps in the US in the last few decades – by the powerful and the regulators to destroy free market principles - it still has a significantly superior system compared to the other two. Since democracy is a necessary condition to fulfill this responsibility, one could safely exclude China from the list, at present. India, under the guise of democracy, has been riding high but one has to raise serious questions about how its economy is organized. If a dozen conglomerates control a large percentage of the GDP, with disproportionate influence on policy and some with a checkered history of legal and regulatory compliance, then it does not have a free enough system to quality. Just having democracy as a social system does not provide any advantage if the economic system is controlled by a few.

Both China and India have thus far resisted floating their currencies – a clear indication that both have a long way to go to understand and appreciate the advantages of free trade. Trade, obviously, has two parts – the physical trade and the financial part. Reduction of tariffs is good on the physical side but an elimination of them is better. But, with a rigged currency, this does not move either country to a regime of free trade. What is worse is that neither seem to understand that distortions created by a non-floating and underappreciated currency lead to incorrect allocation of capital. Labor cost advantages enjoyed by one in cheap manufacturing and the other in cheaper services exist only because of the pegged currencies. This builds up pressure over time and the ultimate and unavoidable result is a sudden and catastrophic revaluation -  with predictable results on the industries that are protected.

It is time for China and India to step up to the world stage – take their share of responsibility to assure that we have a sustainable environment and a stable system. Democracy is a necessary condition for this but not sufficient. An economic system with free trade,  free floating currencies and free markets is equally important. Without these conditions, both India and China are at the risk of squandering their opportunities – one by the shortsightedness and the other by the  incompetence of its leaders.

Macro entanglement

Recent observations that may point to entanglement at larger scales are intriguing as it may aid a framework shift, something that is long overdue. This may degrade classical Physics to one of convenience rather than something with a defendable framework. Human scale observations are convenient for humans. If quantum effects are universal at all scales, then most of the effort currently wasted to empirically prove or disprove expectations based on status-quo, may have to be redirected. If the reason that quantum effects are not observed at macro-scale is that people have not been looking for it, it may have set humans back many decades.

Classical Physics has been enormously soothing for most involved and the human rationality is fundamentally based on it today. We reject and accept hypotheses based on crude and lumpy observations. This also means that little effort is expended to seek the unknown or explain the inexplicable. It has also neatly segmented theories into the micro and the macro, with no real requirement to connect them. Those in one bucket or the other, would rather spend their time going deep in their bucket – never worrying about the implications of what they find for those on the other side. This is efficient scientific research – one that maximizes the number of papers produced and increases the probability of a prize.

If quantum effects are indeed universal, regardless of scale, then one has to seriously question the current state of knowledge. If classical Physics evolved only as a matter of convenience, then it is time to move on.