Wikipedia:Articles for deletion/Hilbert-Hermitian wavelet

From Wikipedia, the free encyclopedia

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was delete. Sr13 09:22, 18 May 2007 (UTC)

[edit] Hilbert-Hermitian wavelet

Hilbert-Hermitian wavelet (edit|talk|history|links|watch|logs) – (View log)

There is no independent discussion on the topic. The Hilbert-Hermitian wavelet is described in the PhD thesis of Jon Harrop and also in some papers by him. There are some references in papers by other people who probably use the wavelet. However, they don't state clearly that they use this wavelet and they don't give any description or analysis of the Hilbert-Hermitian wavelet.

There is much more details and discussion at Talk:Hilbert-Hermitian wavelet (including conflict of interest and accusations of spam and stalking) but the above is in my opinion the salient point. -- Jitse Niesen (talk) 03:42, 13 May 2007 (UTC)

  • The authors of the academic papers did state clearly that they use our product that implements (only) this wavelet. Jon Harrop 04:18, 13 May 2007 (UTC)
    • I didn't find any clear statement in the paper by Christie, Taraskin, and Elliott (arXiv:cond-mat/0406248. The paper by De Nyago Tafen, M. Mitkova and D. A. Drabold indeed says that it uses your product. -- Jitse Niesen (talk) 06:02, 13 May 2007 (UTC)
      • That is exactly what I found too. In both cases the coverage of the wavelet was extremely brief. See Talk:Hilbert-Hermitian wavelet#Notability criteria for a more thorough discussion of the results found. (Requestion 07:57, 13 May 2007 (UTC))
        • Yes. This wavelet started life as something of an academic curiosity only of interest to waveleticians but it is now used in all of the major branches of science by scientists and engineers all over the world. Most of its users do not understand it in detail, they just use it. Conversely, I don't understand what the genomicists in Switzerland do with it, or what the Australians using it to analyse sleep patterns do with it, or what the analysis of heterodyne velocimetry data done by the guys at Los Alamos National Laboratories is all about. I just know that they're all using it. So you're not likely to find a detailed discussion of this theory in their papers. Jon Harrop 09:55, 13 May 2007 (UTC)
      • Christie was our first user. Jon Harrop 09:55, 13 May 2007 (UTC)
    • So Jon, are you saying that your commercial Mathematica notebook CWT product implements only the Hilbert-Hermitian wavelet and nothing else? (Requestion 08:02, 13 May 2007 (UTC))
      • Our time-frequency analysis product computes continuous wavelet transforms using the Hilbert-Hermitian wavelet. It only provides this wavelet. If you want to use other wavelets then you must code them up yourself. Jon Harrop 09:55, 13 May 2007 (UTC)
  • comment. Work has been peer-reviewed (Ph.D. dissertation from Cambridge and several papers), so does not qualify as original research. So as suggested by Jitse, the crux of the issue is "notability". Here it seems to me that being published in peer-reviewed journals and having one's software used by other researchers is sufficient, even if very marginal. I realize this is not as noteworthy as in pure mathematics, as applied guys are always publishing stuff everywhere...:-p; however Jitse's comments on the talk page for the article are very similar to what I just wrote. I would like to know what changed Jitse's mind from "marginal keep, not worth nominating" to "AFD nom". --C S (Talk) 04:49, 13 May 2007 (UTC)
    • I think it was when I said "what if Requestion doesn't know his arse from a quadrature mirror filter?" that tipped the balance for Jitse, who replied "Your comment attacking Requestion was singularly unhelpful and loses you a lot of respect.". So I strongly advise you to not mention Extreme Ironing. Jon Harrop 05:16, 13 May 2007 (UTC)
    • My opinion did indeed change. I used to think "it's in the gray area, not worth nominating", while I now think "weak delete; let's get a decision". I changed from abstain to weak delete because of the extra information that was brought to the talk page. However, I usually don't nominate "weak deletes", so more important is another development: the continuous fighting around the article and related external links. I think it would be good have a wider discussion, get a decision on whether to keep or delete the article and close this chapter. -- Jitse Niesen (talk) 06:02, 13 May 2007 (UTC)
    • Two important quotes from the WP:SCIENCE notability guideline are "unless it generates a significant outside response by the scientific community or the population at large, focusing on such research in a Wikipedia article does not adequately conform to Wikipedia's policy on neutrality, in particular the section on undue weight" and "papers covering the contribution have been widely cited". As discussed for three weeks on the talk page, the cited references do not appear to meet the thresholds set by those two questions. (Requestion 07:23, 13 May 2007 (UTC))
      • Several problems with just quoting WP:SCIENCE like that. First it's just proposed. Secondly, it's generally been agreed by those working on the proposal that "significant" and "widely cited" depends greatly on the specific subfield or discipline. Lastly, this proposal has gotten very little input from mathematics editors, which is particularly significiant given the proposal's qualifiers on the dependence on what is normal in a discipline. I actually participated in this discussion, and some editors agreed they should ask for input from the WikiProjects they intend the proposal to encompass. But they never asked WikiProject Mathematics, so I wonder if this proposal can even be said to include mathematics (which many people do not consider a science). --C S (Talk) 13:44, 13 May 2007 (UTC)
        • True, WP:SCIENCE is a "proposed" guideline. If we choose to ignore WP:SCIENCE then what are the alternatives? Using only WP:NOTABILITY does not seem like a good option to me. The WP:SCIENCE notability guideline is well written, coherent, and it contains a lot of useful wisdom from many notability experts. I agree, since it is only a proposed guideline, that we should not quibble over words like "significant" and "widely" but instead focus on the spirit of what WP:SCIENCE is attempting to accomplish. I'm active over at WP:EL and my experience is that the guidelines are in a constant state of flux; wording is frequently being refined, new sections get added, obsolete ones get refactored or deleted, new issues get resolved. It just part of the Wikipedia guideline / policy making process in action. If no one uses WP:SCIENCE then there will be no motivation to improve it. (Requestion 18:00, 13 May 2007 (UTC))
  • Comment You're basically asking regular editors to decide whether some incredibly, incredibly abstruse (but possibly quite notable) mathematical concept is notable or not. The majority, if not all of, the editors who do regular AfD patrol won't have a clue what a wavelet is, let alone whether this specific form of wavelet is a) real and b) considered notable in the mathematical community. We're lucky if we know how to turn on our calculators. I think you have to submit this to a panel of experts. (PS. Only on Wikipedia can there be a massive flame war *and* edit war about something like this.) --Charlene 07:12, 13 May 2007 (UTC)
    • Yeah, there's no way this is going to work. This article could be one huge hoax, and I would have no idea. --Haemo 07:31, 13 May 2007 (UTC)
    • WP:IAC. We should be able to determine if WP:SCIENCE has been met. (Requestion 07:37, 13 May 2007 (UTC))
To the best of my knowledge, our product is the only product that implements this wavelet. However, all of the research into this wavelet was done in academia and all of the results are publically available accordingly. So you are free to use it. Jon Harrop 10:01, 13 May 2007 (UTC)
  • Delete. According to my understanding of WP:SCIENCE neither " being published in peer-reviewed journals" , nor "having one's software used by other researchers" are enough to establish notability, by a long shot. A couple of citations are not enough to qualify, certainly not in a field like wavelets where truly significant work gathers hundreds of citations (e.g. [1]). Beside this, this article is apparently being used to AFFIRM the importance of the author's work. Here is not the place. Stammer 14:04, 13 May 2007 (UTC)
  • Delete. Ignoring all the side issues, it comes down to demonstration of notability. Computing/physics topics, even obscure ones, are very well-represented on the Web. If this had current notability (even on the academic paper circuit) we'd be seeing more than 100 Google hits. Tearlach 15:12, 13 May 2007 (UTC)
  • Delete I am not sure whether a handful of citations necessarily gives or ought to give N in other areas, but I am sure it does not necessarily do so in science. About 100,000 scientific papers a year have 4 or more citations. To make an individual discovery notable as a subject, whether in a paper or thesis or whatever, takes more than that. For methods work in general, I concur with Tearlach that it takes many more than this, though I do not know the details for this subject--for subjects I do know such a claim would be absurd. Are there other grounds: not as the basis for an add-onto Mathematica--most such are NN. When it becomes a standard method, then it will be notable. It's early days yet. Though the inventor is an expert, entering into the argument to this extent is an example of why we have rules about COI: a person can fairly describe what his work is, but cannot fairly evaluate its importance.
For any non-scientist who's gotten this far, I'd recommend keeping track of the COI--this applies to experts as to ordinary people, and it doesn't take an expert to judge. DGG 03:39, 14 May 2007 (UTC)
  • Delete - Fails WP:SCIENCE. In addition, Jon's statement above that [t]o the best of my knowledge, our product is the only product that implements this wavelet raises serious COI concerns on its own, and this wavelet not being implemented by others futher exacerbates the notability concerns. --EMS | Talk 18:04, 14 May 2007 (UTC)
  • Delete Does not meet WP:NOTABILITY and WP:SCIENCE criteria. Violations of WP:COI, WP:NOT#SOAPBOX, and WP:SPAM. (Requestion 23:15, 14 May 2007 (UTC))
  • Weak delete. If there were even a half-dozen papers by independent groups directly on the topic of this kind of wavelet, I think that might be enough. But one publication cited by three or four independent papers that use this material incidentally (and, seemingly, would be happy to use any other kind of wavelet as long as the code worked) doesn't seem like enough to me, and clearly fails to meet WP:SCIENCE. No prejudice against re-creation if, a few years down the road, this kind of wavelet takes off and picks up hundreds of citations. —David Eppstein 17:56, 15 May 2007 (UTC)
  • Delete without prejudice for recreating once it gets a wider reception. It's a bit sad, because it is an interesting work and one cannot deny the mathematical argument behind this. But this should have no influence on inclusion decisions. Wikipedia isn't the right place to announce that your newly discovered method X has superceded old method Y. Wikipedia will report once this has happened in the field. --Pjacobi 21:52, 15 May 2007 (UTC)
  • Delete There are too few independent mentions of the content, but the icing on my cake is "our product is the only product that implements this wavelet" statement. Let the experts write papers and review the concept, and let those papers and reviews be published. Then the article can be re-created. The COI issue here looms too large for me. KrakatoaKatie 07:30, 16 May 2007 (UTC)
The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.