User talk:JanWMerks

From Wikipedia, the free encyclopedia

Welcome!

Hello, JanWMerks, and welcome to Wikipedia! Thank you for your contributions. I hope you like the place and decide to stay. Here are a few good links for newcomers:

I hope you enjoy editing here and being a Wikipedian! Please # sign your name on talk pages using four tildes (~~~~); this will automatically produce your name and the date. If you need help, check out Wikipedia:Questions, ask me on my talk page, or place {{helpme}} on your talk page and someone will show up shortly to answer your questions. Again, welcome!  —Ruud 00:48, 28 March 2006 (UTC)


Contents

[edit] hello

Hi mr werks. ive read your comments on geostatistics, and found them to be informative. I was not aware of the issues that you raised until you presented them in the kriging and geostats articles. in the quest for scientific and statistical relevance we must always confront inaccuracies. the only comment/suggestion i can make regarding your contributions, is that the antagonistic perspective that you present detracts from the valid points that you make. SCmurky 00:01, 10 June 2006 (UTC)

Hi SCmurky, Thanks for your message. I've been exposed to geostatistical peer review since the early 1990s, which explains the antagonistic perspective of a true contrarian. I may not make a complete recovery any time soon! --Iconoclast 00:47, 23 June 2006 (UTC)

[edit] orphaned article

Hello. Your newly created article spatial dependence was an orphan, i.e. no other articles linked to it. I added it to the list of statistical topics. You might want to consider which other articles should link to it, and add the links. Michael Hardy 22:25, 29 March 2006 (UTC)

Thanks for your note. Linking to statistical topics makes sense but linking to geostatistics makes no sense whatsoever. I have asked geostatistical scholars such as Armstrong, Clark, Journel, Krige, and scores of others, the same fundamental question: does or doesn't each distance-weighted average have its own variance? So far to no avail! Google "Bre-X bogus barren" to find my website, look under Correspondence and peruse what Stanford's Journel wrote about spatial dependence and "Fischerian" [sic] statistics in the early 1990s. I'm proud that Dr F P Agterberg, President, International Association for Mathematical Geology, calls me an iconoclast. --Iconoclast 21:57, 1 April 2006 (UTC)

[edit] spatial dependence

  1.  ::Spatial dependence is a measure for the degree of associative dependence between independently measured values in a temporally or in situ ordered set, determined at different locations in a sample

The above is not an appropriate way to begin an article. In the first place, the typical reader stumbling upon this does not know whether it is about geometry, theology, politics, financial markets, etc., after reading the sentence above. That is one reason why articles often begin by saying "In chemistry, ...." etc. In the second place, you've neglected to highlight the title phrase at its first appearance. And thirdly, there are no links. Michael Hardy 23:55, 1 April 2006 (UTC)

[edit] kriging

I'm afraid you still need to get used to some Wikipedia conventions. Note that the title word or title phrase should normally be in bold at its first appearance in an artilce. Kriging was like that before you edited it. Also, you added material at the beginning without really incorporating it into the article. It's as if a new article starts with the second paragraph, with a completely new introductory sentence. Michael Hardy 01:47, 2 April 2006 (UTC)

I'll keep trying and do appreciate your efforts to make me a Wikipedian. Please be patient. Thanks--Iconoclast 19:17, 2 April 2006 (UTC)

I hope you would be able to explain what kriging is. That would be more helpful than saying that "Geostatisticians refuse to explain why the variance of the distance-weighted average vanished before it turned into a kriged estimate." As it stands, without any reference, it is impossible for us to judge this statement, and such a strong attack requires a lot of evidence. Cheers, Jitse Niesen (talk) 02:45, 3 April 2006 (UTC)
Hello, Jan. I know you're new to Wikipedia, so I'd like to point out WP:NPOV, which is an essential element of Wikipedia culture. Please read it -- it will help you work smoothly with other Wikipedians. The essential element is that Wikipedia should not push a particular point of view: it should merely report on them. If a significant number of statisticians think that geostatistics is broken, please cite them. If only you hold that belief, then it would be original research and is not appropriate for Wikipedia. Please follow these links to understand Wikipedia better.
Since your material is controversial, let's discuss it at the appropriate Talk pages: Talk:Geostatistics and Talk:Kriging.
Thanks! -- hike395 05:03, 3 April 2006 (UTC)

If I believed that scientific integrity is irrelevant on Wikipedia, then I would be wasting my time. Please do not assume that I'm the only one who believes that geostatistics is a scientific fraud. Since the early 1990s I've posed the same question but have yet to receive a single answer; did or didn't each distance-weighted average have its own variance before it was reborn as a kriged estimate? That's all! Thanks and regards,

--Iconoclast 16:03, 3 April 2006 (UTC)

With regard to the Wikipedia's NPOV policy I should mention that geostatistical peer review is a blatantly biased and shamelessly self serving sham. Some evidence is posted on under Correspondence. Much more will be presented in a different format at a later stage. Nothing about the Kriging Game is NPOV! --Iconoclast 16:51, 3 April 2006 (UTC)

Hi, Jan. I'm afraid that your own questioning of Geostatistics is not a verifiable source: in fact, that would be classified as original research. Geostatscam.com is registered to Jan Merks (through GoDaddy.com), so that would also count as original research. Are there any other sources you can cite that shows that the belief that geostatistics is a scientific fraud would be a significant minority belief? -- hike395 04:33, 4 April 2006 (UTC)

Hello Hike 395, Here's what you ought to study between hikes: Look under Documents, play Clark and the Kriging Game by entering coordinates beyond Clark's sample space and watch the distance-weighted average and its variance converge on the arithmetic mean and the Central Limit Theorem. You may also look at the formula for this variance and figure out if it converges on the Central Limit Theorem. One would hope this theorem is not an NPOV subject because it is already posted on Wikipedia. Look at SU19921015 under Correspondence in which the Journelian doctrine that spatial dependence may be assumed is formulated in a letter to the Editor of Mathematical Geology. Please note Journel's non-NPOV remark about Fisher's work. Read JMG19921026 to find out where MG's Editor stands. And peruse JMG20060220 and learn what MG's current Editor-in-Chief said about geostatistics. A synopsis of my case against geostatistics can be found in JMG20030321. And you think I'm the only one who disagrees with assuming, kriging, smoothing and rigging the rules of mathematical statistics? Much of my work can be found in ISO standards such as those developed by ISO Technical Committee 69-Applications of statistical methods. Do Wikipedia's rules of engagement really preclude a critique of a true scientific fraud? Surely, you must be joking! Talk to a friend who scored a passing grade on Statistics 101 and understands what degrees of freedom are all about. Or talk to a professor who teaches mathematical statistics and ask him whether or not each functionally dependent value has its own variance. He would be surprised to find out that this question qualifies as original research but he won't apply for a research grant any time soon. Regards,--Iconoclast 16:26, 5 April 2006 (UTC)

Wikipedia's rules preclude any novel work done by an editor him/herself. That's what we mean by "no original research." We don't mean material that is covered by a research grant or would be suitable for research work in a university. It doesn't matter whether the original work is trying to uncover scientific fraud, or accounting fraud, or high crime and misdemeanors --- we represent all significant sides of the controversy: you just need to verify that it is significant (please see Wikipedia:Verifiability). All of your references to geostatscam.com are not valid Wikipedia references, because you made that web site yourself. The reference to ai-geostats doesn't help, because it is your interpretation of the Kriging game that you're trying to cite. Where, exactly, within the numerous publications of ISO Technical Committee 69 does it say that geostatistics is invalid?
If you'd like to discredit geostatistics, please provide reference(s) which point to a significant number of statisticians who think that geostatistics is globally incorrect. Then, we can write something like "A significant number of statisticians, including XXX and YYY [citation 1][citation 2], think that geostatistics is invalid for the following reasons, blah blah blah."
Again, it doesn't matter that citation 1 or citation 2 are pushing a point of view, or that the geostats community is pushing a point of view. We're just going to follow the pillars of Wikipedia: Wikipedia:NPOV, Wikipedia:No original research, and Wikipedia:Verifiability
-- hike395 02:06, 6 April 2006 (UTC)
P.S., by the way, regarding your comment "Talk to a friend who scored a passing grade on Statistics 101 and understands what degrees of freedom are all about.", please see Wikipedia:No personal attacks.

Hi Hike395, You know plenty about my views on geostatistics and mathematical statistics but I know next to nothing about your views on either subject. What I want to know most of all is why you believe that geostatistics is scientifically sound, and why mathematical statistics is orginal research. After all, we would not want a conflict of interest to taint our views and violate Wikipedian rules, do we? --Iconoclast 15:18, 6 April 2006 (UTC)

The beauty of Wikipedia is that my personal views should not matter. Nor should yours. I have never stated that geostatistics is scientifically sound, nor have I stated that it was dubious, nor a fraud. It doesn't matter -- we just need to produce verifiable content and not do original research. --- hike395 15:28, 7 April 2006 (UTC)

And that's all I do. I post neither questionable content nor original research. Let's leave it at that! --Iconoclast 16:56, 7 April 2006 (UTC)

Hello Custodians of Scientific Integrity, Dr Merks is my son and a lead architect behind EMF (Eclipse Modeling Framework). We took the same stats courses at the same university albeit at different times. We coauthored Precision Estimates for Ore Reserves, a paper based on mathematical statistics but thrashed by geostatistical peer reviewers at CIM Bulletin, Mathematical Geology and IMM’s Transactions before it was praised by and published in Erzmetall see Reviewed papers. In this paper we applied Fisher’s forbidden F-test to prove the presence of spatial dependence between gold grades of in situ ordered rounds in a drift. Fisher’s F-test is the essence of analysis of variance, which is why Fisher became Sir Ronald A Fisher in 1953, the year of the Korean War Armistice. It is preposterous to suggest that applying Fisher’s F-test to verify spatial dependence is somehow “original research”. Analysis of variance is an integral part of internationally recognized standards on sampling and statistics. So we investigated geostatistical theory and practice and determined that it violates the requirement of functional independence and ignores the concept of degrees of freedom. These fundamental and irrefutable facts failed to strike the geostatocracy as particularly alarming. So here are the same facts but in a simplified nutshell:

  1. The arithmetic mean is the central value of a set of measured values with equal weights.
  2. The area-, count-, density-, distance-, length-, mass- or volume-weighted average is the central value of a set of measured values with variable weights.
  3. Each central value is a functionally dependent value of a set of measured values.
  4. Each central value of a set of measured values has its own variance.
  5. Each distance-weighted average is the central value of a set of measured values determined in samples selected at positions with different coordinates in a sample space.
  6. Each set of measured values determined in samples selected at positions with different coordinates a sample space gives an infinite set of distance-weighted averages.
  7. Each distance-weighted average in an infinite set has its own variance.
  8. Each kriged estimate in an infinite set no longer has its own variance.
  9. A set of measured values determined in samples selected at positions with different coordinates gives an infinite set of kriged estimates aka kriged estimators.
  10. The kriging variance of a subset of some infinite set of kriged estimates is an invalid measure for variability, precision and risk.
  11. The kriging covariance of a subset of some infinite set of kriged estimates is an invalid measure for associative dependence.
  12. Spatial dependence between measured values in an ordered set should not be assumed but verified by applying Fisher’s F-test to the variance of the set and the first variance term of the ordered set.
  13. Degrees of freedom are positive integers for set of measured values with equal weights and positive irrationals for set of measured values with variable weights.

Surely, Wikipedia’s policy of NPOV only makes sense if the science behind a discipline does not crumble under scrutiny. Reading my retro reviews of the first three textbooks might assist because studying geostats is not for the faint of heart or the statistically challenged. Study Clark’s Practical Geostatistics first if you are either blessed or cursed with an irresistibly inquisitive mind! And explain how to break bad news on Wikipedia! Kind regards,--Iconoclast 16:15, 10 April 2006 (UTC)

I'm not sure how else to explain the application of Wikipedia policy. Let's break it down into several issues:
  1. Stating that Krige and Matheron knew things and didn't know other things.
  1. Everything in Wikipedia should be verifiable. It's very difficult to verify whether someone knew something or not: you have to refer to an autobiography. I have not seen any citations to an autobiography -- you've just stated things as bare facts. Please look at WP:V --- if there is something that is unverifiable, a Wikipedia editor can delete it.
  1. Claiming that geostatistics is broken because it doesn't use Fisher's F-test and/or assumes spatial dependence a prior.
  1. All we're asking for is some external evidence that Fisher's F-test is used by at least a minority of kriging practitioners. It doesn't help to make a long explanation here why it's a good idea.. That's what we mean by original research. The correctness or incorrectness of an approach is not a useful Wikipedia discussion --- we report on controversies, we don't decide. Take a look at WP:NPOV. Note that we only discuss minority viewpoints that are significant: that means viewpoints that are very uncommon (say, held by only you or your son) are not reported.
  1. For kriging, I went out and researched what statistical tests on spatial dependence were commonly performed, listed them. Note how I edited the article --- Bayesians make assumptions on spatial dependence, frequentists would do tests. If you can supply references to books or papers not written by yourself or your family, that support the use of Fisher's F-test for spatial dependency, I would be more than happy to list it in the list of tests.
We are just going around and around in circles here. I'm happy to stop reverting your edits if you supply non-self-citations to support the edits you are making. Instead of trying to explain why your edits make sense, find supporting evidence outside of Wikipedia, outside of your own work, and outside of your own interpretation of what other people said.
Thanks! -- hike395 18:00, 10 April 2006 (UTC)
Another way of saying this is that Wikipedia is a very poor tool to convince the world of your own views. In fact, it is explicitly out-of-bounds. See Wikipedia:Wikipedia is not, under "Wikipedia is not a soapbox." Because of the Verifiability and NPOV and No original research rules, you are not going to be able to use Wikipedia to convince the world of your views on geostatistics. If you convert a significant minority of practitioners to your viewpoint by means other than Wikipedia, then Wikipedia will report on the controversy.
It's much better to use communication tools such as your own web site or Usenet news groups to try and convince people. Otherwise, it'll just be a very frustrating experience for everyone involved.
- hike395 18:36, 10 April 2006 (UTC)

Kriging derives from the ubiquitous eponym that recognizes Professor D G Krige’s contribution to geostatistics. Krige was the pioneering plotter of distance-weighted averages at the Witwatersrand gold complex in South Africa where he found the infinite set of distance-weighted averages but lost the correspondingly infinite set of variances. Professor Dr G Matheron and his students were so mesmerized by the infinite set of distance-weighted averages that they, too, failed to notice that the correspondingly infinite set of variances had vanished without a trace. Krige’s honorific eponym inspired a cult like lingo of neologisms such as kriged estimate, kriged estimator, kriging variance, kriging covariance, kriging matrix, kriging method, kriging model, kriging plan, kriging process, kriging system, block kriging, co-kriging, disjunctive kriging, linear kriging, ordinary kriging, point kriging, random kriging, regular grid kriging, simple kriging, universal kriging, and last but not least the zombie zero kriging variance and the inane unit kriging covariance. See caution against oversmoothing in CBul198903 under Articles and letters. Is any Wikipedian addicted to the voodoo science of assuming spatial dependence between measured values, interpolating by kriging, selecting the least biased subset of some infite set of kriged estimates, smoothing its pseudo kriging variance to perfection and rigging the fundamental rules of mathematical statistics? --Iconoclast 21:47, 10 April 2006 (UTC)

[edit] We're just going in circles

Dear Dr. Merks: I have a strong feeling we're just going around in circles, wasting each other's time. I've been around Wikipedia for a while -- I support the central tenets of Wikipedia, and believe that they work well. I've been trying to take your edits and adapt them to meet the policies and guidelines. Other users (such as User:Michael Hardy and User:Antandrus) have also been doing helping to do so.

I know you are a newcomer, and I've tried to explain specifically how the various Wikipolicies apply to each case. Is there any way to reach a compromise, where you can edit the articles while still following our guidelines?

You seem very committed to your point of view that geostatistics is a scientific fraud: you've even gone to the effort of making a whole website to explain your reasoning. That's OK -- you are certainly within your rights to try and convince people that kriging is incorrect. But, at Wikipedia, we don't decide controversies, we just report on them. Wikipedia is not a soapbox --- would you be willing to make edits in a more detached NPOV way?

Or, if you find being detached from this controversy too repugnant or difficult, would you consider working on another part of Wikipedia, where you would find NPOV editing to be easier? One possible strategy is to try working on some other articles, learn about Wikipedia, then come back to geostatistics and kriging when you've had more practice? This is just a suggestion -- there is no compulsion to do so.

Anyway, I would propose cooling off and stop our back-and-forth editing. Can we discuss what to do without thrashing many articles?

-- hike395 06:18, 15 April 2006 (UTC)

[edit] Gridlocked

Good griefs, Hike395, I better confess! I wanted to find out fast if Wikipedia has a role to play in my crusade against voodoo statistics. So far, there’s good news and bad news. But first a synopsis of my case against Wikipedia’s NPOV policy:

  1. NPOV 1: geostatistics: each kriged estimate doesn’t have its own variance.
  2. NPOV 2: mathematical statistics: each distance-weighted average has its own variance.

These NPOVs are mutually exclusive so that one is true and the other is false. I have posted a reference to Volk’s Applied Statistics for Engineers, which explains why Fisher’s F-test is the essence of analysis of variance. This textbook also gives a general formula on which the variance of the weighted average is based. I also posted Lipschutz’s Probability, which gives the same formula for the variance of the weighted average as is posted on AI-Geostats Open Website.

I do admire Wikipedians who are dedicated to Wikipedia’s objectives. However, if a Wikipedian were to remove a contribution that refers to analysis of variance, Fisher’s F-test, spatial dependence or sampling variogram, my original contribution will promptly reappear. That’s the good news. Here’s the bad news. Wikipedia will have to deny access to the only mineral sampling expert who determined that geostatistics is a bogus science and did so long before detecting that Bre-X’s Busang was bogus gold deposit. Never assume! Always verify! I'll keep my eye on the Geostatistics and the Kriging knot. Amazon still has preread copies of Volk’s textbook.--Iconoclast 04:02, 16 April 2006 (UTC)

Hello, JanWMerks. I'm not sure whether your comment about "role to play in my crusade" is sarcasm or not. It's hard to tell just from the text. I sure hope it was meant sarcastically, because it goes completely against the purpose of Wikipedia (see Wikipedia:What Wikipedia is not).
It seems that you don't understand the NPOV policy. NPOV can't be a plural. NPOV isn't about unattributed statement of facts. If different sets of people believe contradictory things, (which happens all of the time, believe me), then Wikipedia should read "Group X believes POV1, Group Y believes POV2, which is contradictory to POV1." The "he said/she said" reporting style is the essence of NPOV. I urge you to read Wikipedia:NPOV to find out more.
Saying that you're the only mineral sampling expert who determined that geostatistics is a bogus science actually works against you including your "geostatistics is bogus" information in Wikipedia. I was hoping that you could find references (other than to your own work) that support your POV --- that way, NPOV would kick in and require us to report both sides. As it stands now, we can't even show that a "significant minority" believe in using Fisher's F-test for spatial dependence. (Fisher's F-test is a well-known test, of course). See Wikipedia:NPOV#Undue weight.
It's really quite a serious breach of Wikipedia:Wikiquette that you are unwilling to compromise about this. Jimbo Wales has stated that NPOV is a fundamental policy that's non-negotiable.
For now, I won't remove the material, but label it as POV.
I hope you'll read more about NPOV and can understand where I (and other Wikipedians) are coming from.
with regards, --- hike395

[edit] Please don't remove my comments from Talk pages

Oh, by the way, it's also a serious breach of Wikiquette to remove or alter other people's comments from Talk pages, as you did to my comments on Talk:Tolstoy syndrome and Talk:Kriging. Please refrain from doing this in the future.. (Notice how I am not editing your comments on Talk pages, even though I am on the main pages?) --- hike395 11:32, 16 April 2006 (UTC)

[edit] How to decide between true and false NPOVs?

That I'll do my way! Please delete whatever does not meet Wikipedia's non-negotiable rules and I shall post it again as long as possible. Beating around the bush is not my style.--Iconoclast 17:47, 16 April 2006 (UTC)

[edit] Variogram

Hi Hike395, I wonder whether or not variogram meets Wikipedia's requirements for style and tone. It looks a rather rambling definition of sorts!--Iconoclast 02:29, 2 June 2006 (UTC)

[edit] Image copyright problem with Image:Sampling_variogram.jpg

Thanks for uploading Image:Sampling_variogram.jpg. The image has been identified as not specifying the copyright status of the image, which is required by Wikipedia's policy on images. If you don't indicate the copyright status of the image on the image's description page, using an appropriate copyright tag, it may be deleted some time in the next seven days. If you have uploaded other images, please verify that you have provided copyright information for them as well.

For more information on using images, see the following pages:

This is an automated notice by OrphanBot. For assistance on the image use policy, see Wikipedia:Media copyright questions. 15:00, 18 May 2006 (UTC)

[edit] Consensus Science

Why do you keep vandalizing the consensus science article? I realize you have a beef with geostatistics, but isn't consensus science. --SpinyNorman 23:59, 17 June 2006 (UTC)

Hi SpinyNorman, Geostatistics is junk science by consensus. In my opinion, it would makes a unique example about consensus science. After all, the geostatistical fraternity have made a mockery of mathematical statistics and peer review.--Iconoclast 01:05, 19 June 2006 (UTC)

Hi JanWMerks, what I see is a content dispute rather than vandalism. However, please discuss your edits on the talk page of the article rather than getting into a revert war over the content. Edit warring tends to lead to Wikipedia:3RR violations. Discussion toward solving disputes is the wiki way. Cheers, Vsmith 01:27, 18 June 2006 (UTC)

Hi VSmith, Shortly, I'll post on this page what I would prefer to post under Consensus Science, Examples. Please give logical reasons why it would be out-of-place.--Iconoclast 01:05, 19 June 2006 (UTC)

Hi SpinyNorman and VSmith, Here's my take on geostatistics: Geostatistics is a unique case of junk science by consensus. Professor D G Krige discovered at the Witwatersrand complex in South Africa that two or more gold assays, determined in samples selected at positions with different coordinates in a finite sample space, define an infinite set of distance-weighted averages. Professor Dr G Matheron (1930-2000) was so impressed by Krige’s aptitude in augmenting sparse data that he conferred the ubiquitous krige eponym. Neither Krige nor Matheron knew that one-to-one correspondence between distance-weighted averages and variances is sine qua non in mathematical statistics. On the contrary, true variances of distance-weighted averages were replaced with false kriging variances of sets of kriged estimates. Thus, geostatistics violates the requirement of functional independence and ignores the concept of degrees of freedom. Geostatistics turned into junk science by consensus when its practitioners made a mockery of peer review. I look forward to feedback.--Iconoclast 16:55, 19 June 2006 (UTC)

[edit] Junk science

Please stop inserting original research into the article. Please refrain from undoing other people's edits repeatedly. If you continue, you may be blocked from editing Wikipedia under the three-revert rule, which states that nobody may revert an article to a previous version more than three times in 24 hours. (Note: this also means editing the page to reinsert an old edit. If the effect of your actions is to revert back, it qualifies as a revert.) Thank you. Thank you. --Howrealisreal 17:11, 22 June 2006 (UTC)

Howdy Howrealisreal, Thanks for letting me know that functional dependence and degrees of freedom qualifies as orginal research in your line of work. Visit my website, click Spreadsheet templates, and look for Clark and the Kriging Game. Play the game by entering coordinates beyond Clark's sample space and watch what happens to the distance-weighted average and its vanished variance. How about that! Enter the coordinates of the moon and look at all that hypothetical uranium in space! Take a look at Fisher's F-test for spatial dependence. There's no significant degree of spatial dependence!!! Of course, Clark is allowed to interpolate because the Journelian doctrine dictates that spatial dependence may be assumed. That's what geostatistics is all about! Assume, krige, smooth, rig the rules of mathematical statistics and be happy!!! I wish I could have inserted a happy face! Kind regards, --Iconoclast 22:06, 22 June 2006 (UTC)

Jan, Wikipedia is not a place for self-promotion. This is an encyclopedia, not an annex for your personal website. If you can source your contributions using scientific journal articles, books, etc... (not created by you) then perhaps you can find a way to word what you want to say that doesn't conflict with the original research rules. You can't just write things from your website here, because then it appears that those views are endorsed by Wikipedia. You see what I'm saying? Thanks for writing back to me.

P.S. - It also doesn't really make sense to document geostatistics extensively in the junk science article. The better home for your analysis (once you get sources) would be in the article specifically dealing with that topic. --Howrealisreal 14:12, 23 June 2006 (UTC)

Jeremy, Since when is promotion of scientific integrity and knowledge synonymous with self-promotion on Wikipedia. I shall continue to post general questions that you may or may not remove. Meanwhile, I suggest you try to find the correct answers to each of these questions. Look at Geostatistics and find out how Wikipedians are trying to reconcile irreconcilable differences. Fortunately, everybody is learning in the process. Kind regards,--Iconoclast 15:38, 23 June 2006 (UTC)

Jeremy, You should not remove any statement the significance of which is not abundantly clear to you. Do you understand that every distance-weighted average has its own variance because it is a functionally dependent value (calculated value!!!) of a set of messured values? Do you understand that functionally dependent values do not give degrees of freedom but that measured values do? In the simplest of terms, measured values give degrees of freedom but calculated values are not similarly blessed. Infinite sets of distance-weighted averages are the equivalent of perpetual motion in data acquisition. It may seem very pretty to the statistically challenged but it is junk science by consensus just the same. Kind regards, --Iconoclast 17:56, 27 June 2006 (UTC)

Jan, I think you are confused. I have not changed the content of the junk science article since June 22nd, at 12:12. [1] If you have a problem with your edits continuing to be removed, please talk to ones that are doing it. --Howrealisreal 18:32, 27 June 2006 (UTC)

Jeremy, Thanks for letting me know that it was not you who removed my most recent explanation why geostatistics is junk science. Geostatistical gurus like to keep under wrap the formula for the long-lost variance of the distance-weighted average. Kind regards, --Iconoclast 22:16, 27 June 2006 (UTC)

---

Jan, thanks for the note on my talk page re junk science. I did do a certain amount of mathematical modelling and statistics as part of my degree, but that was a very long time ago, so I class myself as not having sufficient understanding to judge whether your position on geostatistics is valid.

However, I think junk science is a much broader subject that should not have as its examples details of relatively minor instances of possible junk science that are impenetrable to most readers of the article and possibly faulty as well. It just confuses matters. That's why I have been removing your contributions from the article.

I've seen other campaigns on Wikipedia that push a particular controversial and weakly-supported line. Generally these just fail; sometimes it takes weeks, and sometimes months. You're on to a loser if you think you'll get anywhere on Wikipedia in the way you have been going. Perhaps you should try getting a science journalist on some prominent media outlet interested, if you haven't already. It sounds like something they might be interested in.

Cheers, --A bit iffy 07:24, 30 June 2006 (UTC)

[edit] Spam?

Please do not add commercial links (or links to your own private websites) to Wikipedia. Wikipedia is not a vehicle for advertising or a mere collection of external links. You are, however, encouraged to add content instead of links to the encyclopedia. If you feel the link should be added to the article please discuss it on the article's talk page rather than re-adding it. See the welcome page to learn more about Wikipedia. Thanks. Vsmith 22:05, 24 June 2006 (UTC)

My non-profit website provides nothing but essential information for the statistically dysfunctional who believe that it makes scientific sense to assume spatial dependence, interpolate by kriging, smooth voodoo variances to perfection and rig the rules of mathematical statistics. For example, my website gives the formula for the variance of the single distance-weighted average that went missing on Krige's watch while he was plotting gold values in the Witwatersrand reef complex in South Africa. And you know what? Matheron and his nascent geostatisticians didn't even notice that this variance vanished! Kind regards,--Iconoclast 22:20, 24 June 2006 (UTC)

Thank you for experimenting with the page Junk science on Wikipedia. Your test worked, and has been reverted or removed. Please use the sandbox for any other tests you want to do. Take a look at the welcome page if you would like to learn more about contributing to our encyclopedia. The Prophet Wizard of the Crayon Cake {Prophesize) 22:01, 25 June 2006 (UTC)

Hello Cake Wizard, You prove that your understanding of mathematical statistics leaves a lot to be desired. Perhaps you should belatedly follow the advice of H G Wells who said that statistical thinking will one day be as necessary as reading and writing. The wizard appears to be a few crayons short of a full box!!!--Iconoclast 22:51, 25 June 2006 (UTC) Please read WP:NPA.Bridesmill 05:07, 30 June 2006 (UTC)

[edit] Bre-X

Hi Jan!

I've reverted your changes to the Bre-X article under the no original research policy. If you want to cite the opinions aired on geostatscam.com, please provide references to show that they are shared by a significant minority. Nvj 09:52, 15 July 2006 (UTC)

Howdy Nvj,

Sir Ronald A Fisher was knighted in 1953 because of his work on analysis of variance. Since then ANOVA has been accepted by a statistically significant majority of scientists. In contrast, a small minority of geostatistically blessed souls replaced the variance of a functionally dependent value with the pseudo variance of a set of functionally dependent values. One of those might be tempted to revert my reference to geostatistical software under Bre-X. JWM.--Iconoclast 18:39, 15 July 2006 (UTC)

Howdy
I'm not casting doubt on whether ANOVA is well-referenced enough, it seems that it indeed is. The problem is that your specific application of it to the Bre-X case has not been shown to be shared by a significant minority. I encourage you to read the no original research policy, specifically the following passage:
An edit counts as original research if it proposes ideas or arguments. That is, if it does any of the following:
[...]
* It introduces an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor, without attributing that analysis or synthesis to a reputable source;
Since the article currently do not have any such attributions I've tagged it with the [original research?] template. Since you know the field so well I'm sure you can find references of the right kind if they're around. If the article seems to stay indefinitely without these references I might delete the tagged paragraphs again. Nvj 10:28, 17 July 2006 (UTC)

Hello, Thanks for your message. Please look at spatial dependence and peruse Journel's letter to JMG's Editor. Perhaps you should also read what JMG's Editor wrote to me! Fisher's F-test for spatial dependence could have nipped Bre-X in the bud but Journel is troubled by classical "Fischerian" (sic!) statistics. The F-test is a precursor to the sampling variogram, a graph that shows where spatial dependence in a sample space dissipates into randomness, and that is incorporated in several ISO Standards. Original research? JWM.--Iconoclast 16:44, 17 July 2006 (UTC)

You're welcome. I read Journel's letter as published on your page. I didn't go through the formulas, as a quick skimming confirmed that it didn't touch on the Bre-X case at all. If you draw conclusions in that regard from this letter they are indeed "an analysis or synthesis of established facts, ideas, opinions, or arguments in a way that builds a particular case favored by the editor". Spatial dependence does not talk about the Bre-X case, and in any case it is largely written by you which makes it invalid for this purpose.
What references do you have where someone else than you apply the statistical techniques you espouse to the Busang deposit? Nvj 18:24, 17 July 2006 (UTC)

First you talk about original research and I point out that there's nothing original about ANOVA and sampling variograms. Then you want to know whether somebody else did the same work as I did on Bre-X. Surely, you must be joking! Of course, nobody did! I could have proved that Bre-X was a salting scam based on duplicate bogus assays for the first three to five borehole. That's why I'm writing a book about Bre-X! I trust you're not a closet geostatistician. JWM. --Iconoclast 22:56, 17 July 2006 (UTC)

Ok, let me rephrase that last sentence: Do you have ANY references to other people sharing your views on how ANOVA etc. applies to Bre-X? (And why haven't you just furnished them already?) To repeat myself: I'm not casting doubt on whether ANOVA is well-referenced enough, it seems that it indeed is. The problem is that your specific application of it to the Bre-X case has not been shown to be shared by a significant minority.
You can remedy that last problem right away by giving appropriate references. -- Nvj 17:37, 18 July 2006 (UTC)

Thanks, Nvj. Here's more than I really wanted to disclose at this stage. Bre-X, Gold Today Gone Tomorrow, gives a summary account of my work for Barrick Gold because my confidentiality agreement lasted until Dec 1999. I wrote Applied Statistics in Mineral Exploration and Boreholes Statistics with Spreadsheet Software without any reference to Bre-X. Both papers were reviewed and published, and are posted on my website under Reviewed papers. The reviewers expected that my views would open a hornet’s nest among geostatisticians but nothing happened. In fact, even though the editor extended the deadline for criticism, the geostatocracy has yet to respond. At some point, the term geostatistical software in my Bre-X contribution was replaced with something like manipulation of data but most mining professionals even know whose geostatistical software was applied. JWM.--Iconoclast 15:53, 19 July 2006 (UTC)

So, what does "Gold today" say about it? Does it actually discuss ANOVA? (Something I wouldn't expect from a "popular" work). Do you have any relevant quotes?

You're correct! Danielson and Whyte's "Bre-X: "Gold Today Gone Tomorrow" mentions ANOVA but doesn't discuss it. Please peruse Danielson's 1998 Editorial in The Northern Miner titled "Challenging geostatistics", which is posted under Articles and letters on my website. Also posted are Letters to TNM's Editor submitted by Canada's most dedicated enforcers of geostatistics. JWM. --Iconoclast 22:14, 25 July 2006 (UTC)

Howdy Nvj, Do you have enough info to infer I didn't engage in "original research" when I applied ANOVA to prove that the intrinsic variance of Busang's phantom gold was statistically identical to zero? Is it you or me who reverts the note on the Bre-X page? Thanks and regards, JWM.--Iconoclast 16:17, 9 August 2006 (UTC)

I've put the original research tag back in, because there is still no reference cited on the page. The expression "original research" means that a Wikipedian has done the research and is attempting to publish it on Wikipedia as a primary medium of publication. To resolve this problem, you need to show that someone has credibly done this research (with respect to Bre-X in particular), that it has been published outside of Wikipedia, and that it has the support of the vast majority of people who are knowledgeable in this area—if it is not a vast majority who support this conclusion, you will need to qualify your statements in the article appropriately, in addition to citing a reference. In any case, unless the conclusion is supported by a significant minority at least, it is probably not good material for an encyclopedia. --Dglynch 01:44, 22 August 2006 (UTC)

I redirected analysis of variance, which dates back to R A Fisher's work in the early 1930s. Need I mention Fisher's F-test? Isn't that the quintessence of ANOVA? There's not a shred of original research in applying Fisher's F-test to verify whether two variances in a measurement chain or hierarchy are statistiscally identical or differ significantly. So I reverted! JWM. --Iconoclast 15:54, 22 August 2006 (UTC)

Please, if you truly believe that this is not original research, replace the OR tag with a proper citation of a reference instead of simply removing it. --Dglynch 16:32, 22 August 2006 (UTC)

[edit] Insulting to Geographic Information Science

ive not seen one person argue with your view statistically speaking, however you have basically besieged an entire field of research with your views. Nvj, are you a statistician, neither am i; mr merks noone is accusing you of being wrong... get it. learn how to write an academic article perhaps. I highly dislike your edits of spatial dependence for example, you draw the discussion of a valid field of statistics, geography, topology, and topography, and turn it into an argument against kriging. Im sure that there is a conspiracy group somewhere that you may participate in, eg. 'I think it is all because of the freemasons'. however geography is a serious science, the applications of which im certain that you have not truly studied, or understood. currently I work for the government in the analysis and creation of geographic information, and personally, I do not use kriging in the limited analysis that i do perform, I prefer the triangulated irregular network or the distance weighted average due to their ease. while i can appreciate your argument on kriging, i do not believe that it belongs in the majority of articles concerning GIScience related topics, as that is what geostatistics is, and while you may feel that certain applications assume spatial dependence, in many cases kriging is simply a method to examine trends, not unlike the difference in elevation and slope between a mountain and its valley.SCmurky 01:07, 19 July 2006 (UTC)

No, Mr. Merks should not "learn how to write an academic article" (and if he doesn't know, it's irrelevant). The matter at hand is writing an encyclopedia entry. An important difference between these two is that original research is a requirement for an academic article (ok, maybe not for certain review articles) while it is frowned upon in an encyclopedia. I hope you can see the difference. Certain skills on the other hand transfer very well from academic article-writing into encyclopedia-writing. Nvj 11:14, 19 July 2006 (UTC)

Howdy SCmurky, I looked in my crystal ball and saw only positive outcomes. I would be very pleased if the prefix "geo" in "geography" had created a variant of geostatistics different from Matheronian geostatistics in the sense that it neither violates the requirement of functional dependence nor ignores the concept of degrees of freedom. However, any references to the seminal work of Sichel, Krige and Matheron tend to make me nervous! Even if your earlier work were based on assuming spatial dependence, interpolating by kriging, and smoothing pseudo kriging variances of set of functionally dependent kriged estimates to perfection, you could still position yourself at the leading edge of commonsensical sampling practices and sound statistical methods in geography, topology and topography by looking at mathematical statistics with a dispassionate eye. So do not engage in artificial indignation but try to make a difference in the science of geography. JWM.--Iconoclast 16:35, 19 July 2006 (UTC)