Talk:Nuclear power/ArchiveOldberg

From Wikipedia, the free encyclopedia

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Removed from the section on risks:

However, their argument rings hollow, for they have failed to address a situation in which the risk is not defined but seems to be (Oldberg, 2005).

This is not NPOV because, clearly, there are many proponents of nuclear power who do not think their arguments on risk are "hollow". You can say "opponents of nuclear power" or "Fred Bloggs says that...". Secondly, what the hell is the actual argument made by the reference? It's completely unclear. --Robert Merkel 07:33, 28 December 2005 (UTC)

Also removed:

According to Oldberg and Christensen (1995) and Oldberg (2005), the implication that "Probabilistic" Risk Assessment is probabilistic is invalidated by empirical violations of probability theory which have been persistently ignored by proponents.

From what I can tell, Oldberg seems to a bit of a lone iconoclast whose views have largely been ignored. While the Google test is not infallible, nobody else seems to link to or cites the papers you've cited. --Robert Merkel 09:57, 28 December 2005 (UTC)

Merkel's argument that "Oldberg seems to be a bit of a lone iconoclast" is ad hominem, non-peer-reviewed and unresponsive to the topic. On the other hand, my assertions are responsive and have been published in the peer-reviewed literature of science. Hence, Merkel's argument lacks standing under Wikipedia's rules of evidence. Why, then, did Merkel feel it proper to delete my posting?

The Nuclear Regulatory Commission admits in their inspection manual that many of the probabilities involved are approximations based on the subjective assessment of the inspector, and that these highly approximate probabilities are multiplied to obtain a result. Is that not consistent with the Oldberg and Christensen reports?

The issue of whether probability values cited in nuclear power industry reports are estimates stands apart from the issue of whether these "probabilities" are probabilities. The several, peer reviewed articles that I have published since 1990 point out that the "probabilities" are sometimes not probabilities. As "risk" is measured by probability, the fact that the "probabilities" associated with the reliability of the defect detection tests of nuclear power are generally not probabilities is quite significant regarding the potentiality for risk assessment and the utility of nuclear power.

Thus far, the Nuclear Regulatory Commission has not admitted or addressed the problem of violations of probability theory in the defect detection tests of nuclear power. This is true, though: a) I identified the problem to the NRC while serving on the advisory committee to an NRC research projects in 1984 b) I identified the disastrous effects of this problem on the same research project in a peer reviewed article published in 1995 c) the NRC appealed the decision to publish and had its appeal rejected by referees d) the NRC has had 21 year in which to frame a refutation to my arguments and has not done so to the satisfaction of peer reviewers.

Have you checked the Thompson Science Ciation Index (SCISEARCH)? I would feel much more comfortable if you would please do so, or ask others to do so here on talk, before removing citations to peer-reviewed sources.

Finally, why did you not copy the citation which you also removed here to talk? Please copy it here. —James S. 22:43, 29 December 2005 (UTC)

Works cited:

Oldberg, T. and R. Christensen, 1995, "Erratic Measure" in NDE for the Energy Industry 1995; The American Society of Mechanical Engineers, New York, NY. Republished by ndt.net at http://www.ndt.net/article/v04n05/oldberg/oldberg.htm

Oldberg, T., "An Ethical Problem in the Statistics of Defect Detection Test Reliability," 2005, Speech to the Golden Gate Chapter of the American Society for Nondestructive Testing. Published by ndt.net at http://www.ndt.net/article/v10n05/oldberg/oldberg.htm

Again, please check SCISEARCH (or at least Google Scholar) before striking that kind of thing. —:James S. 00:55, 30 December 2005 (UTC)

Just because the NRC says something, and Mr. Oldberg also says something similar, that the NRC is even aware of Mr. Oldberg's work. I can't find a single piece of evidence that they, or anybody else, have given it attention. I did check Google Scholar already (sorry for not making this point explicit) - I would have checked Web of Knowledge but I don't have access to it over the holiday season. In any case, there's not one single mention on Google Scholar. Oh, and while the first reference appears to be peer-reviewed, an invited talk generally doesn't count as such. --Robert Merkel 06:39, 30 December 2005 (UTC)

Great, thanks; I didn't mean to imply otherwise.

In the mean time, can we talk about what the actual effect of multiplying so many uncertainties implies in terms of the mathematical 95% confidence interval? I remember seeing an amortization of the cost of waste disposal that someone had written out in the mid-1980s, and I just can't see the current state of the art having advanced all that much. —James S. 08:11, 30 December 2005 (UTC)

The statement that the NRC may not be aware of "Mr. Oldberg's work" reveals that UTC has not read the references on which he is commenting. Were he to read them, he would find that the NRC is a) aware of the work b) has disputed the finding c) had its position rejected by peer reviewers. Regarding which of the works cited were peer reviewed, my 1995 paper was published under the rules of peer review of the American Society of Mechanical Engineers; the U.S. Nuclear Regulatory Commission disputed the decision to publish and had its position rejected. An abstract of my 2005 presentation was submitted to a chapter of the American Society for Non-destructive Testing and was accepted for presentation by the organization's board of directors. Following presentation, I submitted the presentation for publication to the online journal nde.net. It reviewed the paper under its rules of peer review and accepted it for publication. The same journal, ndt.net, has vetted my 1995 paper's claims over a period of more than 6 years in its online forum. No refutation or limitation of the claims have been presented. By the way, I published a third, peer reviewed article on the same theme circa 1990 in the peer reviewed journal Materials Evaluation; it claims that steam generator tube inspection methods fail to define statistical populations.

Then someone should put your comments back in the article. —James S. 07:13, 3 January 2006 (UTC)

Firstly, I was not aware that the individual adding the references to Oldberg's work was in fact Oldberg. Frankly, on the evidence available to me I don't think a description as a "lone iconoclast" is inaccurate or an ad hominem attack. For the benefit of others who are reading this, the claim that something has been published in a peer-reviewed journal doesn't automatically make it a) correct or b) so significant that it should automatically be mentioned in Wikipedia. Many, many thousands of scientfic papers are published each and every year; most will go largely unnoticed. For the conclusions of a paper to be significant and appropriate to mention in Wikipedia, the work has to have significantly influenced thinking on the topic; generally, in the context of scientific papers this will be when *other* works cite the paper; it might also be other books, government reports, and even in the case of nuclear safety if Greenpeace or some other prominent anti-nuclear group used the work to support their campaign that would be evidence that the work has been influential. However, as far as I can tell the work hasn't been cited by anyone else, anywhere. If you can provide evidence that it has been influential in some substantial way, well and good. If not, your work shouldn't be mentioned in Wikipedia. --Robert Merkel 08:02, 3 January 2006 (UTC)

Robert Merkel: The description of me as a "lone iconoclast" is an example of what philosophers call an "ad hominem argument." It is one of the logical fallicies; they are so named because they can make a false conclusion seem true. To make an ad hominem argument is to make one's opponent in debate the issue rather than an assertion of one's opponent. You have surely made an ad hominem argument, for whether there is an error in nuclear power reactor safety inspection systems is what is at issue but you have made me the issue.Terry Oldberg 20:58, 8 January 2006 (UTC)

Actually, if the NRC appeals process is as Oldberg has described, then his work has been cited favorably in government documents. Being cited in peer-reviewed literature isn't necessary to show veracity. Of course it helps, but on controversial topics the truth often -- perhaps more often -- comes out through court or law enforcement investigation, not the peer-reviewed literature. In some cases university labs will try to get there first, but step back and think about what we are talking about. There is already a staff of dozens at the NRC, in Congress, and as various claims work their way through the courts, there are even professional expert witnesses who are thinking about these things. Even the private sector has actuaries who have to figure out what to charge in various underwritings. Oldberg's thesis is mathematically sound, because any introduction of subjective judgement magnifies uncertainty. If all references uncited by the peer-reviewed literature were excluded, I doubt that would be an improvement. —James S. 08:24, 3 January 2006 (UTC)

It doesn't necessarily have to be cited by peer-reviewed literature, but find somebody who has given this research serious attention, be it in a court of law, in a legislative body, in the scientific/engineering literature. Unless substantial attention has been paid to it by a relevant group of people, it doesn't matter how brilliant it is, it's not for Wikipedia at this point in time. --Robert Merkel 09:09, 3 January 2006 (UTC)

I agree that just about anyone can get published in a peer reviewed journal. Some journals even make money by making authors pay money to be printed. I think something should have been verified by numeruous studies before it comes in. This is the sort of thing that ends up leading to ~1/3 the conclusions from medical stuidies being retracted and the laymans scepticism of scientific experts. If you want people to ignore wikipedia, put in lots of stuff that is cutting edge research and may eventually be proven incorrect. The previous statements are general, I haven't been keeping up with the exact discussion above. Lcolson 13:55, 3 January 2006 (UTC)

Robert Merkel and Lcolson: You've misrepresented the system by which truths are established in science. This system is not litigation in a court of law, a vote in a legislative body or having attention paid to it by a relevant group of people. The system is peer reviewed publication. That a paper has been published under peer review signifies that: a) reviewers, who are expert in the field, have reviewed it b) the reviewers have found no errors in the logic or the facts that are presented c) the reviewers feel the topic is important enough for the paper to occupy the limited number of pages that are available in the journal. Once a paper has been published, the claims that are made in it become open for review and possible refutation within the scientific community. If a refutation is not forthcoming, the paper's claims stand. The mechanism for correcting a faulty claim is not, as you seem to think, to ignore it. It is to publish a refutation of the claim. I've published three, peer-reviewed articles on the topic of statistical anomalies in the field of defect detection testing. The first, published in the February 1991 edition of Materials Evaluation, a journal of the American Society for Nondestructive Testing, claims that inspection methods for PWR steam generator tubes do not define statistical populations. This claim has not been refuted in the subsequent 14 years. The second, published by the American Society of Mechanical Engineers, claims that defect detection tests generate empirical violations of probability theory; hence they are inconsistent with conventional statistics. In 1995, Business Week magazine printed about a million copies of an article about this paper. The paper was the topic of an article in the French magazine La Reserche (sp?). I forwarded copies of it to the U.S. Nuclear Regulator Commission, the U.S. Federal Aviation Agency, the C.E.O.s of all nuclear utilities in the United States, the C.E.O.s of all U.S. reactor vendors and numerous academic scientists. Six years ago, the Web based journal ndt.net republished the article. It has been available for discussion and has been discussed in the forum of ndt.net; ndt.net has an international readership of 80,000 specialists in nondestructive testing. The paper has also been available for discussion and discussed in the online forum of the American Society for Nondestructive Testing. On two occasions, I've presented the paper to live audiences of experts in nondestructive testing. I've been a paid consultant on the paper to the U.S. Nuclear Regulatory Commission. Despite this vast amount of exposure, despite the great incentive that proponents of nuclear power have to refute the paper, and despite the fact that it has been in print for more than 11 years, it has not been refuted. This paper claims to invalidate research by the U.S. Nuclear Regulatory Commission on the reliability of PWR steam generator tube inspection methods. The NRC disputed this finding and had its position rejected by the peer review system. The scientists who had their research panned by this paper have not published a refutation. Is the thesis of my papers established within the scientific community? Yes it is. Is calling attention to this thesis contrary to the interests of promoters of nuclear power? Yes it is. Is the topic of incompetent engineering of the safety inspection systems of nuclear reactors of sufficient importance to be included in Wikipedia's article on nuclear power? It surely is. Terry Oldberg 20:21, 8 January 2006 (UTC)

I met with Terry Oldberg and doubt any editor reading this would consider him "just about anyone." Among other things, he was a LLNL nuclear weapon designer with a "Q" security clearance from 1963-1968. The following comprise his peer-reviewed and invited work (NDT=nondestructive testing):

“An Ethical Problem in the Statistics of Defect Detection Test Reliability,” Address to the Golden Gate Chapter of the American Society for Non-destructive Testing, March 10, 2005.

“Erratic Measure,” NDE for the Energy Industry 1995, pp. 1-6. The American Society of Mechanical Engineers, New York, NY (1995).

“Concept of Concreteness and Abstraction in Estimating NDT Reliability,” Materials Evaluation, 49, pp. 297-299 (1991).

“Fission Gas Release and Fuel Reliability at Extended Burnup: Predictions by the SPEAR-BETA Code,” Meeting on LWR Extended Burnup - Fuel Performance and Utilization, Williamsburg, VA, April 4-8, 1982.

“Probabilistic Code Development,” Entropy Minimax Sourcebook, Vol. 4, p 29. Entropy Limited (1981).

“New Code Development, “ ibid, p 35.

“Spear Code Development,” ibid, p 45.

“Entropy Minimax Hazard Axes for Failure Analysis,” ibid. p 674.

“Advances in Understanding and Predicting Inelastic Deformation in Zircaloy,” Zirconium in the Nuclear Industry, American Society for Testing and Materials (1980).

“The Carnot Efficiency Principle as a Guide to Decision Making in Research,” Invited Lecture, Stanford University, circa 1980.

“Dealing with Uncertainty in Fuel Rod Modeling,” Nuclear Technology, 37, p. 40.

“Thermographic Imaging of Nuclear Fuel Rods,” Nuclear Technology (1978).

“Non-Steady-State Factors in Models for Swelling of Oxide Fuels,” Nuclear Applications and Technology, 9, p. 338 (1970).

“Important Mechanisms in the Explanation of Clad Diametral Expansion at First Startup of a Mixed Oxide Fuel Pin,” Proceedings Conference on Fast Reactor Fuel Technology, New Orleans, April 13-15, 1971. American Nuclear Society.

“Behave-2: An Oxide Fuel Pin Performance model in Two Spatial Dimensions and Time,” Transactions of the American Nuclear Society (1971).

“Mechanical Model of Thermal Differential Expansion at First Startup of a Mixed Oxide Fuel Pin,” Transactions of the American Nuclear Society, 13, p 573 (1970).

I think his views are proper to include. —James S. 04:54, 8 January 2006 (UTC)

Well that is fascinating (there's lots of questions I'd love to ask Mr (Dr?) Oldberg, virtually all of which he wouldn't be permitted to answer and to which he'd probably have to dumb down the answers for me to comprehend), but it still doesn't alter the fact that none of the above provides any evidence that other people are paying attention to his views expressed the two papers being cited. See WP:NPOVUW.--Robert Merkel 12:33, 8 January 2006 (UTC)

The statement "Such probabilistic assessments have been criticized by authorities because they lack mathematical rigor and involve subjective assessments" quotes unnamed "authorities" - you'll have to cite a source for them. In addition, in 2005 Oldberg himself said that his 1995 paper was being ignored (after skimming it, I can understand that - he obviously knows nothing about steam generators or water chemistry in nuclear power plants). I'm removing the sections involved (can't revert - there's some renewable stuff in the way). Simesa 17:46, 8 January 2006 (UTC)

Simesa: You're wrong when you say "...he obviously knows nothing about steam generators...". From 1982 to 1886, I was one of about 6 people who managed the 30 million US$ research project of an international group of owners of pressurized water reactors calling itself the "Steam Generator Owners Group." Among my responsibilities was management of research and development on methods of steam generator tube inspection. In the same period, I sat on the steering committee of the U.S. Nuclear Regulatory Commission's research program on steam generators, a large portion of which was devoted to an attempt at measuring the reliability of inspections. I'm a co-author of "The Steam Generator Reference Book." Electric Power Research Institute, 1995. Terry Oldberg 20:33, 8 January 2006 (UTC)

Oldberg and Christensen are the authorities in this case. What does nondestructive testing of dry cooling pipes have to do with water chemistry? The fact that the pressurized cooling pipes corrode at unpredictable rates is not in dispute. The fact that the means of testing them for corrosion is not mathematically rigorous (as explained by Oldberg and Christensen 1995) is peer-reviewed, and has not been answered in the literature. Please do not remove the assertions unless doing so is supported by peer-reviewed sources. Furthermore, the comment that punped hydro power storage is "prohibitatively expensive" is nonsense. Pumped hydro sotrage has been in use for almost 100 years, and at the lowest efficiency configurations barely doubles the resulting cost. —James S. 19:23, 8 January 2006 (UTC)

I stand corrected - I should have read Oldberg's website, not just the references and the list of publications above. Water chemistry in a nuclear power plant tells you how many tubes are already leaking - they know about what they're going to find before they do the inspections. Exact location would seem to be unimportant - a sufficient indication anywhere on the tube causes it to get plugged (plug enough tubes and you have to de-rate the plant). Oldberg is one authority and should be cited explicitly, not in as generic a statement as "the authorities" (implying all qualified engineers).

Prohibitively-expensive, or something like it, is justified. The reference in this section says:

In a March 2004 report Eurelectric and the Federation of Industrial Energy Consumers in Europe pointed out that "Introducing renewable energy unavoidably leads to higher electricity prices. Not only are production costs substantially higher than for conventional energy, but in the case of intermittent energy sources like wind energy, grid extensions and additional balancing and back-up capacity to ensure security of supply imply costs which add considerably to the end price for the final consumer." "Reducing CO2 by promoting renewable energy can thus become extremely expensive for consumers," though both organisations fully support renewables in principle.

Note that pumped storage absorbs about 25% of the energy it can produce (see Pumped-storage hydroelectricity

I ask pardon for not checking earlier. I'm involved in college and an arbitration, and could use a wikibreak.

Simesa 22:54, 8 January 2006 (UTC)

Oldberg "implications"

Removed original research not in Oldberg's peer-reviewed paper. Removed POV assertion that NRC and ASME ignored Oldberg. Wrote to NRC and requested their position on Oldberg's work. Simesa 18:58, 19 February 2006 (UTC)