Gill Eapen's Blog, page 65

Un-Identified

News (1) that researchers from Ruhr-Universität Bochum have been successful in circumventing SSO (Single Sign On) systems, deploying SAML (Security Assertion Markup Language), using techniques, that can be easily learned and replicated, is troubling. SSO was offered as a convenient solution by many systems and SAML has been touted as the standard of security.

Fixing these vulnerabilities is not going to be easy. This can crimp the growth of internet commerce as the general public grow increasingly weary of broken promises of the behemoths. From a risk perspective, there are two important attributes to consider – the probability of a loss of information and the severity of such a loss. Unfortunately, convenience and security are becoming orthogonal, providing direct trade-offs – higher convenience (SSO, for example) resulting in lower security.

Innovation in security has been lagging – with many of the large companies who could have a significant impact on it – taking a path to tactical profits at the cost of consumer confidence. Perhaps, it is too much to ask for in a myopic world of profits and bonuses. Those dominating search would rather strip and store any available information, those dominating operating systems would rather incorporate information gathering agents into systems and those dominating social channels shun security as something that does not matter. We have seen this before in other once high-flying and now lagging industries such as pharmaceuticals and investment banking. And, the ending of this movie will not be any different for internet moguls.

If the leaders of the industry do not step forward to provide the public with a higher level of confidence in internet transactions, they will ultimately lose.

(1) Be whoever you want to be! Published: Wednesday, August 15, 2012 - 10:41 in Mathematics & Economics. Source: Ruhr-Universitaet-Bochum

Stress Tax

Recent finding (1) that stress and depression can shrink the brain of humans has implications for organizational design and policy. Stress acts to reduce the capabilities and performance of organizations and societies. Thus, any agent that causes stress should be taxed appropriately. This is not unlike a pollution tax. Since the stress causing agent imposes a cost on the organization and society, she is part off, such a cost should be mediated by a tax to optimize the system.

Societies without artificial constraints such as those imposed by religions and ad-hoc regulations are likely to have lower aggregate stress than those controlled by a few people and by prescriptive rules. In the case of companies, bureaucratic and hierarchical structures with leadership incompetence on top can cause significant stress in the entire system with disastrous consequences (2). In the case of societies – religious and regulatory rigidities have demonstrably shown to be sub-optimizing as they drive the participants to higher stress and lower performance. Autocratic regimes are extreme examples of regulation in which production and consumption decisions are governed by defined rules.

Stress – the cause of disease and incompetence in humans – may be eating the core out of modern societies. Those causing it should be taxed accordingly.

(1) Yale team discovers how stress and depression can shrink the brain. Published: Sunday, August 12, 2012 - 15:32 in Health & Medicine. Source: Yale University

(2) Flexibility : Flexible Companies for the Uncertain World. Gill Eapen. http://tinyurl.com/doflex

Non-scientific economist

In a podcast (1) an economist seems to argue that economics is not science because it is unable to make precise quantitative predictions of outcomes. When asked if Biology is any better, he responded that such precise predictions are just round the corner in that field. It is unclear what these new developments are.

He seems to be lost in the forest while looking to measure the height of growing trees and not being able to do it. The idea that predictability is a necessary condition to qualify a field to be scientific is exactly what is dampening innovation in those fields dealing with complex non-linear systems such as high energy Physics, Biology and Economics. For example, in financial economics, deterministic bias and the desire for predictability have fully rejected pursuits that attempt to systematically consider uncertainty and flexibility as fundamental drivers of behavior. The same biases have led Biology and Pharmacology to dead ends in drug discovery and high-energy Physics to the particle zoo and machine building to prove irrelevance.

By inculcating an expectation of determinism and predictability as necessary for “true science”, educators are arresting the development of new ideas in these areas.

(1) Rosenberg on the Nature of Economics. Alex Rosenberg. Hosted by Russ Roberts. http://www.econtalk.org/archives/2011/09/rosenberg_on_th.html

Big diagnosis

A recent study (1) demonstrates that the presence of thyroid disease can be predicted with high accuracy by analyzing symptoms, if a large enough data set is available. Medicine has lagged in the use of information in early diagnosis for many different reasons. First, doctors after many years of education and practice are unwilling to give into analytics and the use of computers in diagnostics. They believe that they have won the right to decide who has the disease and who does not – and technology simply cannot help. Second, a system of misaligned incentives among payers, providers and users does not value early diagnosis and prevention sufficiently highly. And, finally, manufacturers have every incentive to slow down diagnosis as such a process has negative economic outcomes for them.  

However, logic and common-sense have to prevail. If information is hiding in the data collected by providers and payers that will aid early diagnosis and prevention, it will be a crime not to use it appropriately. If providers can get over their ego, payers, their incompetence and manufacturers, their greed, we can deploy currently available technologies to fundamentally change healthcare. It will take both creativity and the willingness to change – but nothing more. Data has been accumulating over the past two decades in low cost storage bins and much of it have not been used in the prediction and the prevention of disease.

Any system that slows down the application of technology to the benefit of all, is simply wrong.

(1) Detecting thyroid disease by computer. Published: Thursday, August 2, 2012 - 11:08 in Health & Medicine. Source: Inderscience Publishers

Crooked Market Hypothesis (CMH)

Many have faulted the Efficient Market Hypothesis (EMH) – an elegant and simple explanation of how markets work – for the financial collapse and the failure of regulatory regimes. In the last several years, many economists have argued against the EMH and in many cases against capitalism itself. Many have set out to form more complex theories, including the analysis of human behavior and psychology in a valiant effort to explain what was observed.

It may not be that complex. EMH proposes how markets should work when everybody plays by the rules and when optimal regulations are present to manage any market failures. Implicit in this assumption is the uniformity of regulations and rules for every market participant, regardless of size, shape and color. Just relaxing this one assumption is sufficient to explain everything that has transpired in the last decade. More importantly, EMH never envisioned a situation in which regulatory regimes will be set up to specifically create market failures – to provide advantages to subsets of market participants.

Those engaged in mind bending research in human behavior to explain it all, may want to consider simpler explanations.

BELLA, BELLA – what have you done?

News (1) that the laser system of Berkeley Lab Laser Accelerator (BELLA) delivered a petawatt of power in 40 femtoseconds in a laser pulse did not get much press, perhaps because it shrunk the 2 mile long accelerator at Stanford Linear Accelerator Center to a 1 meter long desktop instrument at Berkeley. If anybody is watching it, it should send strong signals to those nurturing visions of bigger tunnels and fancier atom smashing in search of more elusive particles.

Thinking small has not been a competence of the current crop of scientists. The elegance of theory aided by equally elegant experiments has been lost in the euphoria of heavy duty engineering. If Physics has to advance to truly new science, it has to embrace new technologies and not just attempt to scale status-quo understanding. By cultivating a culture of linear scale thinking, contemporary science has significantly reduced the probability of fundamental shifts.

Creativity will always trump linear scaling – in any domain.

(1) BELLA laser achieves world record power at 1 pulse per second.

Published: Friday, July 27, 2012 - 13:05 in Physics & Chemistry

Source: DOE/Lawrence Berkeley National Laboratory

Tail risk

Scientists and activists alike have been lamenting about global warming and warning the general public about the impending doom (1). This is fair, for sure - no sane animal would burn fossil fuels uncontrollably so as to choke everybody to death. What is not so clear is the prioritization of the risks faced by humans. Research interests of scientists appear to be highly correlated with the ability to publish papers when data is available in plenty.

However, it is important to put this in perspective. Take a little known (or studied) risk for example – the risk of a magnetic polarity change for the Earth. Magnetism trapped in ancient pottery shows that over the past 4,000 years, the magnetic field has weakened by more than 50%. This past century, the strength has dropped by 5% (2) . At this rate, the field might disappear in the next few hundred or thousand years. Another warning sign of an imminent flip has come recently from satellite measurements of the Earth's magnetic field. The implications of a polarity switch can only be imagined – as it is currently “not funded,” to be researched. Another area, that is considered less “sexy,” compared to ideas of imminent death by green house gases ( that includes Methane from cows), is death by an approaching meteor (3). Of course, one can never underestimate the mighty human with his stockpiles of nuclear weapons and accumulating nuclear waste from producing “clean energy,” to do something irrational that will make such risks moot.

Sure, climate change is a concern – but the probability of catastrophic events, that will make sea level changes, massive hurricanes and disappearing ice cover, child’s play are tangible enough to make the tactical concerns less relevant. Tenure and publications, however, follow fashionable research.

(1) Climate concerns. Published: Thursday, July 26, 2012 - 15:35 in Earth & Climate. Source: Harvard University.

(2) http://www.guardian.co.uk/science/story/0,3605,748510,00.html.

(3) Asteroid With Chance of Hitting Earth in 2029 Now Being Watched 'Very Carefully'. http://www.space.com/622-asteroid-cha...

Money for nothing & Options for free?

The implicit guarantee of a bailout provided by the Fed and the US treasury to large financial institutions is a put option written by them (and by the US taxpayers) and provided for free (currently) to these institutions. It allows the financial institution to walk away from its obligations and put the bank to the public for the bailout in a bad scenario. This option does not have any value in a good scenario where the bank makes money and the economy prospers.

This is an extremely valuable option – which I am sure, some of the financial gurus will be able to value if they put their minds to it. if not, a simple Black-Scholes calculator can help. More importantly, it is time that this transfer of value from the taxpayers to the institutions is stated in the national accounts. It is one thing to write an option for free but it is quite another to consider the option to be cost-free. The former is a policy question and the later a common-place accounting issue.

There are no free options in this world except for this. Perhaps the concept of no-free lunch has not sunk in yet for the saviors of the economy and the world.

Too big and fail?

A recent study (1) shows that fastest growing cities are not the most prosperous. Business leaders may want to take inspiration from this finding as they incorporate size and growth into the goals of every executive. Growth is important, for sure, but growth for the sake of it or objectives that focuses on becoming “big,” are not necessarily good. In the 90s, the leaders of a high growth pharmaceutical company (2) set out to become the largest company in the world. They did and in the process destroyed $400 Billion in shareholder value. Contemporary information and internet companies are on a similar path to doom as they dabble in every field in which they have no competence in.

Countries and cities have to focus on per capita metrics rather than aggregate numbers as a way to measure progress. It is not growth of the population or size, that is important – just the opposite. Fast growth in aggregate numbers may indicate weakness in many different dimensions including infrastructure, health and education. As the world’s fastest growing and highest trade surplus country presides over the largest and most painful migrations of life in and out of cities, it has to ponder the implications of it all. Just as large companies, large cities and countries are likely pushing themselves to the brink of disaster. If the growth is engineered by monopoly power, currency manipulations or subsidies, such failures will be catastrophic.

Stable growth with a happy, healthy and educated population is a necessary condition for long term survival – for companies, cities and countries.

(1) New study finds fastest-growing cities not the most prosperous. Published: Thursday, July 19, 2012 - 13:07 in Mathematics & Economics. Source: SAGE Publications

(2) Flexibility : Flexible Companies for the Uncertain World. http://www.amazon.com/Flexibility-Flexible-Companies-Uncertain-World/dp/1439816328/ref=sr_1_3?ie=UTF8&qid=1342925184&sr=8-3&keywords=gill+eapen

Bit to bus

Recent development (1) that showcases a switch that connects a qubit with its bus, increases the probability of practical quantum computing within our life-time. This is an important area that requires more investment and focus. If it does happen, it will qualify as the quantum jump in technology since computers themselves. Practical quantum computing will significantly accelerate knowledge gain in many areas including life sciences, astrophysics and climatology.

A fundamental leap in computing has cascading beneficial effects to society. It can drive humanity closer to solving the nagging problems faced by nearly 6 billion people – problems stemming from lack of food, housing, education and overall health. Designer molecules can stem diseases and improve agricultural productivity. Better prediction of natural disasters and impending climate changes can both reduce catastrophic loss of life and lead to better design of habitats. Better designer materials can improve housing and power generation and improve communication and transportation networks, resulting in higher mobility, education and integration.

Allocation of public and private funds to research have to consider the ability to make systemic impacts to society. In a world mired in tactics, this is probably an unrealistic goal.

NIST develops 'dimmer switch' for superconducting quantum computing. Published: Wednesday, April 28, 2010 - 09:53 in Physics & Chemistry. Source: National Institute of Standards and Technology (NIST)