Softpanorama
(slightly skeptical)
Open Source Software Educational Society
May the source be with you, but remember the KISS principle ;-)
Softpanorama Search
|
Softpanorama |
May the source be with you, but remember the KISS principle ;-)
Softpanorama Search
|
| News | Recommended books | Recommended Links | Deception | Lysenkoism and politization of science | IT offshoring Skeptic |
Pseudo Economics |
Groupthink |
Supply side fairy tales, by Steve Waldman: Greg Mankiw offers a strong endorsement of a proposal to cut the corporate income tax from 35 to 25 percent, claiming "It is perhaps the best simple recipe for promoting long-run growth in American living standards." ... A good case can be made for cutting or even eliminating the corporate income tax. But Mankiw's argument does not cohere.
Let's start positive. Mankiw is right to point out that the "incidence" of the corporate income tax might not in fact be as progressive as its proponents would wish. He quotes studies suggesting that workers end up paying 70% to 92% of the taxes in the form of lower wages. I'm skeptical of those numbers, but it is surely true that some fraction, perhaps even a large fraction, of the corporate tax burden falls on workers and customers rather than presumptively wealthier investors. Mankiw does us all a service by reminding us of this.
Then he tells us a fairy tale ...
... ... ...
Supply side economics is a nice story, a hopeful story. It offers a clean, plausible policy framework: encourage investment, always and everywhere, and prosperity is sure to follow. But this decade has been about a pure a test of that idea as we could hope for. Capital in the United States was incredibly cheap, and what did we do? We destroyed a lot of wealth. We don't need more capital (although we might soon, if our foreign backers get skittish). We need more discriminating capital. In the meantime, the only thing I'm sure "works" about the supply side story is that it shifts the tax burden from richer to poorer. I'd rather that stop working so well.
See also discussion Economist's View Supply-Side Fairy Tales
the lack of risk-taking is particularly evident in computer science:what strikes me is how much pomp, circumstance and apparatus academia requires in order to frame even a very small and simple point. References to everything in the literature ever said on any vaguely related topic, detailed comparisons of your work to whatever it is the average journal referee is likely to find important -- blah, blah, blah, blah, blah.... A point that I would more naturally get across in five pages of clear and simple text winds up being a thirty page paper!
I'm writing some books describing the Novamente AI system -- one of them, 600 pages of text, was just submitted to a publisher. The other two, about 300 and 200 pages respectively, should be submitted later this year. Writing these books took a really long time but they are only semi-technical books, and they don't follow all the rules of academic writing -- for instance, the whole 600 page book has a reference list no longer than I've seen on many 50-page academic papers, which is because I only referenced the works I actually used in writing the book, rather than every relevant book or paper ever written. I estimate that to turn these books into academic papers would require me to write about 60 papers. To sculpt a paper out of text from the book would probably take me 2-7 days of writing work, depending on the particular case. So it would be at least a full year of work, probably two full years of work, to write publishable academic papers on the material in these books!
Furthermore, if as a computer scientist you develop a new algorithm intended to solve real problems that you have identified as important for some purpose (say, AI), you will probably have trouble publishing this algorithm unless you spend time comparing it to other algorithms in terms of its performance on very easy "toy problems" that other researchers have used in their papers. Never mind if the performance of an algorithm on toy problems bears no resemblance to its performance on real problems. Solving a unique problem that no one has thought of before is much less impressive to academic referees than getting a 2% better solution to some standard "toy problem." As a result, the whole computer science literature (and the academic AI literature in particular) is full of algorithms that are entirely useless except for their good performance on the simple "toy" test problems that are popular with journal referees....his first scenario makes me wonder if amateur scientists could again make meaningful contributions to research, combined with a wiki-like process that (hopefully) would identify promising directions better than today's peer reviews:
And so, those of us who want to advance knowledge rapidly are stuck in a bind. Either generate new knowledge quickly and don't bother to ram it through the publication mill ... or, generate new knowledge at the rate that's acceptable in academia, and spend half your time wording things politically and looking up references and doing comparative analyzes rather than doing truly productive creative research.
This classic paper is also available from UMBC CSEE Department and many other places on the Internet.
IMHO Lysenkoism is the term that describes the same phenomenon but in much more dangerous and potent variation. Lysenkoism probably can be defined as cult-style "cargo cult science" where leaders of the cult understand that their methods are false of completely inadequate, but still continue to pursue the chosen course due to specific for large organizations social constrains on their behavior. Unfortunately the term "Lysenkoism" is not that well known and rarely understood in the USA. But IMHO its time to understand the danger it represents for the science and technology.
...With the Science Gravy Train coming to the end of its line, its passengers are falling over each other to find other rationalizations for their craft. In cargo-cult fashion, new threats to national security are created to perpetuate a cold-war mentality (the "war on drugs") and to justify scientific programs ("the asteroids are coming"). Such schemes prey upon our fears and go unchallenged because, in an environment of scarce funding, it is more difficult to apply critical, skeptical, scientific thinking to the source of one's own livelihood. Good science tends to be defined as "science that someone will pay for."
Feynman used the cargo-cult parable to point out certain practices in science. People go through scientific rituals, employ expensive apparatus, and write their journal papers, but when you examine carefully what they're doing, you sometimes find that they've really been fooling themselves. They have set out to produce a certain result, or they forget to ask certain critical questions about what they're doing or why. Or they may even find that questions about the premises underlying their research are forbidden by those who pay for it, and that certain results are deemed unacceptable in advance. When this happens, scientists are just being used. Feynman called this cargo cult science because people are just imitating scientific behavior. An example is industry-sponsored "research" on the health effects of tobacco. Sometimes the consequences are disastrous, as in the Challenger explosion. How much post-cold-war science has this flavor to it?
The author also provides a list of a very interesting question that illustrates the problems mentioned above. Among them:
5. Does a scarcity of science funding create a more competitive environment among scientists that discourages traditional scientific openness and cooperation? Scientists with whom I used to openly share ideas are now pumping me for free data and information to put into their proposals, while at the same time withholding information about their own discoveries. Scientific meetings have turned into cautious discussions of where the money is. I know scientists who are spending more time writing proposals than on science. Just as sports are no longer about sport, science is becoming less and less about science and more about programmatic alphabet soup. This is what happens when scientists spend more time thinking about survival than about science. The science community needs to look into these tendencies for science to become as competitive and exclusive (and occasionally, greedy) as the business world. What changes in the way science is managed could restore an atmosphere of openness and cooperation in a declining science job market?
6. Does a scarcity of support for science make us take a shorter term view of the payoffs we demand of our science? Does it place less value or credibility on basic research and curiosity-driven science? If so, then the scientific community is not doing its job of convincing those who fund science (ultimately, the public) of the value of long-term, basic, and curiosity-driven research. In making a case for the value of science to society, the scientific community often falls into the trap of listing past examples of high-tech consumer fallout from scientific efforts whose original purpose was something completely unrelated. The public must understand that science is not about non-stick cookware and microwave ovens, but about finding out how nature works. This often requires major long-term, unwavering commitments. Science, like education, is an investment in the future.
Is the critical, skeptical thinking that lies at the very heart of scientific inquiry being bred out of the institutionalized science being done today? 7. Does a less secure job environment for scientists make scientists more or less critical of the scientific merit of their own work and of scientific management policies? When I ask scientists why they are doing what they are doing, the answers are often dismaying: "Because my boss told me to; because that is what the sponsor is paying me to do; this is the only thing I could get funding for." Questioning a project's scientific purpose, goals, and procedures is increasingly suppressed, for fear of "biting the hand that feeds us." More cargo-cult science! Is the critical, skeptical thinking that lies at the very heart of scientific inquiry being bred out of the institutionalized science being done today? Left alone, such a trend could cripple scientific inquiry.
We may be entering an era in which only the very best can make a living doing science. 8. Would you encourage your child to pursue a career in science? If capable children are discouraged from pursuing careers in science, because it is no longer a viable profession, the future of science itself is in jeopardy. If you want your kids to go into science, they ought to believe that (a) such a career would be rewarding, not frustrating, and (b) that they would stand some chance of making a reasonable living doing what they are trained to do. In the New Era, both are likely to be true only if you're a very good scientist. We may be entering an era in which only the very best can make a living doing science -- something like the situation we now have in sports and the arts. That's why it would be a mistake to address fears about a scarcity of adequately trained scientists in the future simply by cranking out more new scientists. What we need is not more scientists, but better scientists.
I find it useful to draw a contrast between two different organizational development styles: "process-oriented" and "commitment-oriented" development. Process-oriented development achieves its effectiveness through skillful planning, use of carefully defined processes, efficient use of available time, and skillfull application of software engineering best practices. This style of development succeeds because the organization that uses it is constantly improving. Even if its early attempts are ineffective, steady attention to process means each successive attempt will work better than the previous attempt.
Commitment-oriented development goes by several names including "hero-oriented development" and "individual empowerment." Commitment-oriented organizations are characterized by hiring the best possible people, asking them for total commitment to their projects, empowering them with nearly complete autonomy, motivating them to an extreme degree, and then seeing that they work 60, 80, or 100 hours a week until the project is finished. Commitment-oriented development derives its potency from its tremendous motivational ability—study after study has found that individual motivation is by far the largest single contributor to productivity. Developers make voluntary, personal commitments to the projects they work on, and they often go to extraordinary lengths to make their projects succeed.
Organizational Imposters
When used knowledgeably, either development style can produce high quality software economically and quickly. But both development styles have pathological lookalikes that don’t work nearly as well, and that can be difficult to distinguish from the genuine articles.
The process-imposter organization bases its practices on a slavish devotion to process for process’s sake. These organizations look at process-oriented organizations such as NASA’s Software Engineering Laboratory and IBM’s former Federal Systems Division. They observe that those organizations generate lots of documents and hold frequent meetings. They conclude that if they generate an equivalent number of documents and hold a comparable number of meetings they will be similarly successful. If they generate more documentation and hold more meetings, they will be even more successful! But they don’t understand that the documentation and the meetings are not responsible for the success; they are the side effects of a few specific effective processes. We call these organizations bureaucratic because they put the form of software processes above the substance. Their misuse of process is demotivating, which hurts productivity. And they’re not very enjoyable to work for.
The commitment-imposter organization focuses primarily on motivating people to work long hours. These organizations look at successful companies like Microsoft; observe that they generate very little documentation; offer stock options to their employees; and then require them to work mountains of overtime. They conclude that if they, too, minimize documentation, offer stock options, and require extensive overtime, they will be successful. The less documentation and the more overtime, the better! But these organizations miss the fact that Microsoft and other successful commitment-oriented companies don’t require overtime. They hire people who love to create software. They team these people with other people who love to create software just as much as they do. They provide lavish organizational support and rewards for creating software. And then they turn them loose. The natural outcome is that software developers and managers choose to work long hours voluntarily. Imposter organizations confuse the effect (long hours) with the cause (high motivation). We call the imposter organizations sweatshops because they emphasize working hard rather than working smart, and they tend to be chaotic and ineffective. They’re not very enjoyable to work for either.
Cargo Cult Software EngineeringAt first glance, these two kinds of imposter organizations appear to be exact opposites. One is incredibly bureaucratic, and the other is incredibly chaotic. But one key similarity is actually more important than their superficial differences. Neither is very effective, and the reason is that neither understands what really makes its projects succeed or fail. They go through the motions of looking like effective organizations that are stylistically similar. But without any real understanding of why the practices work, they are essentially just sticking pieces of bamboo in their ears and hoping their projects will land safely. Many of their projects end up crashing because these are just two different varieties of cargo cult software engineering, similar in their lack of understanding of what makes software projects work.
Cargo cult software engineering is easy to identify. Cargo cult software engineers justify their practices by saying, "We’ve always done it this way in the past," or "our company standards require us to do it this way"—even when those ways make no sense. They refuse to acknowledge the tradeoffs involved in either process-oriented or commitment-oriented development. Both have strengths and weaknesses. When presented with more effective, new practices, cargo cult software engineers prefer to stay in their wooden huts of familiar, comfortable and-not-necessarily-effective work habits. "Doing the same thing again and again and expecting different results is a sign of insanity," the old saying goes. It’s also a sign of cargo cult software engineering.
But the fundamental problem was structural: the government were creating a managerial structure that separated the power to influence treatment from clinical responsibility for the consequences of that treatment. For example, managers were claiming the right to determine the nature of a drug prescription while doctors remained both morally and legally responsible for the outcome. This seemed self-evidently unethical, and placed doctors in the impossible position of being vulnerable to sacking of they were disobedient and to malpractice suits if they obeyed. The system was also open to corruption, since political influence could be brought to bear on managers, tending to generate protocols using criteria dictated by expediency rather than effectiveness. For example, the call for specified protocols on psychotropic drugs were unsupported by any rational consensus on what such protocols should contain, and the instructions concerning ▒skin▓ cancer (and the failure to differentiate between basal cell, squamous cell and melanomatous malignancies) seemed more justified by the immediate demands of public relations than by any body of solid scientific evidence.
As things turn out, I had underestimated the seriousness of this kind of threat to clinical practice, and my article unfortunately proved to be prophetic of a trend which has culminated in the creation of NICE and CHI. The management of science in medicine is now established by statute, and - even worse - the criteria of effectiveness have been conflated with economic considerations of "cost-effectiveness". The stage is set for clinical science to be steamrollered by the demands of power politics.
A style of (incompetent) programming dominated by ritual inclusion of code or program structures that serve no real purpose. A cargo cult programmer will usually explain the extra code as a way of working around some bug encountered in the past, but usually neither the bug nor the reason the code apparently avoided the bug was ever fully understood (compare shotgun debugging, voodoo programming).
The term "cargo cult" is a reference to aboriginal religions that grew up in the South Pacific after World War II. The practices of these cults centre on building elaborate mockups of aeroplanes and military style landing strips in the hope of bringing the return of the god-like aeroplanes that brought such marvelous cargo during the war. Hackish usage probably derives from Richard Feynman's characterisation of certain practices as "cargo cult science" in his book "Surely You're Joking, Mr. Feynman" (W. W. Norton & Co, New York 1985, ISBN 0-393-01921-7).
Richard P. Feynman, the late Nobel laureate in physics, stressed the importance that scientists not fool themselves by referring to the cargo cult people of the South Pacific after the war (Feynman, 1985). These aboriginal islanders wanted to make U.S. cargo planes return with all kinds of goods, so they erected towers and wooden antennas near the airstrip, acted like controllers, and waited for the planes to come in. Their form was correct but no planes came in. He calls this "cargo cult science," where you do all the right things, you think, but you are wrong, nevertheless. You either leave something out or draw the wrong conclusion. What is missing, Feynman says, is "utter scientific integrity," meaning "a kind of utter honesty, a kind of leaning over backwards," the duty "to report everything you think might make your conclusion invalid," and "giving details that could throw doubt on your interpretation." It's this type of integrity, this care not to fool yourself, that he says is missing in much of the research in cargo cult science. He gives examples of investigators fudging data not fitting the theory they wanted to prove. "The first principle is that you must not fool yourself — and you are the easiest person to fool," he says.
Feynman's first principle applies to any type of important investigation. In child abuse cases the absence of investigative integrity reduces the process to cargo cult medicine and law. Law and medicine rely on each other to such a degree that each suffers from the investigative flaws of the other in these cases. These flaws include improper belief systems or biases, institutional pressures, carelessness, and lack of proper training. Doctors and social workers in the medical system claim they are not investigators. However, the legal system often takes action based on what they said and did with the child before the police entered the picture, and on the conclusions they draw.
Copyright © 1996-2009 by Dr. Nikolai Bezroukov. www.softpanorama.org was created as a service to the UN Sustainable Development Networking Programme (SDNP) in the author free time. Submit comments This document is an industrial compilation designed and created exclusively for educational use and is placed under the copyright of the Open Content License(OPL). Site uses AdSense so you need to be aware of Google privacy policy. Original materials copyright belong to respective owners. Quotes are made for educational purposes only in compliance with the fair use doctrine.
Disclaimer:
Last modified: August 10, 2009