Jump to content

Wikipedia:Articles for deletion/Non-universality in computation

Add links
From Wikipedia, the free encyclopedia
< Wikipedia:Articles for deletion
This is an old revision of this page, as edited by Arthur Rubin (talk | contribs) at 16:01, 24 June 2007 (→[[Non-universality in computation]]: comment). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Non-universality in computation

Non-universality in computation (edit | talk | history | protect | delete | links | watch | logs | views) – (View log)

Non-notable concept, used only by its inventor. An additional issue with the article is that the meanings of the terms used deviate from the common meanings used in the field of computability theory, making the article confusing and misleading. Discussion with the author (see Talk:Non-universality in computation) has failed to resolve this.  --LambiamTalk 12:47, 20 June 2007 (UTC)[reply]

  • Delete per nom. Also, all the sources are from one author (a couple have dual authorship, but still the same author appears there as well). What's up with that? –King Bee (τ • γ) 13:07, 20 June 2007 (UTC)[reply]
  • Delete per nom. Seems like a case of promotion of a crank/unestablished idea using Wikipedia. --Cronholm144 13:51, 20 June 2007 (UTC)[reply]
  • Delete. Article on concept with single proponent created and supported by single-topic contributor so looks very much like vanity article. No independent sources. Gandalf61 14:03, 20 June 2007 (UTC)[reply]
  • Weak delete. Isn't this just an overblown restatement of the time hierarchy theorem? On the other hand, I think merging there would be counterproductive. —David Eppstein 15:49, 20 June 2007 (UTC)[reply]
  • Delete. A brief look at a technical report by Akl suggests he can express himself well in the vocabulary of conventional computer science. If he were the one writing a Wikipedia article on his views, he might be able to do a better job than the article we are now considering for deletion. He has created some new subtleties around the edges of the Turing Machine definition, at points where the TM assumes something 'just happens' with zero effort. However, even in that paper the actual importance of the new subtleties is not clear. If his idea catches the general attention, and other people write about it, that will be the time to consider having a WP article about his theories. EdJohnston 15:58, 20 June 2007 (UTC)[reply]
  • Delete this. Akl seems to me to be arguing that a Turing machine cannot operate in real time, or on quantities, like quantum measurements, affected by the computation; yet these are still "computations" which a Universal Computer should be able to do. He is not "disproving" the Church-Turing thesis, he is denying it. A note to this effect may be useful at that article, so I will copy the links there. This article fails to make any of that clear. Nevertheless, he is cited by other people (some of them his co-authors, but that's not uncommon) so he should be mentioned in that article. Septentrionalis PMAnderson 17:33, 20 June 2007 (UTC)[reply]
  • Do NOT Delete. Hello everyone, so I am the author of the page and just want to start by apologizing for the abundance of references by the same author. To be totally honest, I was in the middle of finding more sources as i was writing the article, however, I was first exposed to Dr. Akl's research so I naturally used him as a reference. I have to stress that this is not a vanity page and i truly apologize if that's how it has come across. I also feel bad for not making the article more user friendly. This was the first full page I have created and struggled with how much technical / non-technical information to present. If given the chance by this committee I would gladly re write the article in as much detail as you all see fit. I am sad that it has gotten to this point where the article is on the chopping block since I truly believe that this kind of research should be on Wikipedia. I have updated the references section of the page and have removed all but one of Dr. Akl's papers. I am in the process of adding another independent reference but i first need permission of the author. I will address each of the comments above in your respective talk pages since some of what was said isn't true. However, judging on the initial ground for deletion, the fact that there was too many sources by the same author and that it was not notable, I have fixed this and will continue to fix it in the coming weeks. As i said before... this is not Fringe research, however, there is no doubt that the research is controversial. I feel that this article and topic represents an exciting area of theoretical computer science and believe that scientists young and old, as well as anyone interested could learn a lot from it. So I apologize if the article was not well written, but I will change the article and have already begun cleaning up the references. For these reasons I do not believe this article should be deleted. Thank you very much for your efforts. If you have any questions or comments please feel free to let me know. ewakened 20:48, 20 June 2007 (UTC)[reply]
    • The new references are that well-known hoax, the Perspex machine, and a paper, of which I have not seen more than the abstract. The abstract warns It is suggested that claims to have "broken the Turing barrier" could be toned down and that the important and well-founded rôle of Turing computability in the mathematical sciences stands unchallenged. I suggest that ewakened take this to heart. I would be content to see an article which did not have a section on "The myth of the Universal Machine"; in the meanwhile, at least userify until ewakened has understood his sources. A redirect and mention at hypercomputation would also be a good replacement. Septentrionalis PMAnderson 22:06, 20 June 2007 (UTC)[reply]
      Comment. See also Wikipedia:Articles for deletion/Perspex machine.  --LambiamTalk 22:31, 20 June 2007 (UTC)[reply]
  • Delete. Although the Perspex machine is not exactly a hoax, it has nothing to do with computability theory. It may have something to do with philopsphy of computability theory, but, even so, the "proofs" here are wrong and unsalvagable. — Arthur Rubin | (talk) 00:00, 21 June 2007 (UTC)[reply]
  • Keep and clean-up - Most of the delete arguments above seem to rely on a novel interpretation of "independent sources." First, the subject is not the author, and second, the references in question seem to have been published by reliable third parties with editorial and/or peer review. The fact that the same author appears on them only means caution must be used, per WP:SPS. (Note to closing admin, please review relevant policy and discount any arguments made contrary to policy). I don't know if this theory is true or not, and neither do the others making arguments that it is not, which is WP:OR anyway. We don't allow inclusion of OR in WP, and we shouldn't allow exclusion based on OR either. Even if a counter-proof was published in a reliable source, we would simply merge that into this article, give each appropriate weight per WP:NPOV, and let the reader decide: "The threshold for inclusion in Wikipedia is verifiability, not truth", per: WP:V. There seems to have been a productive discussion on the talk page for this article, and the normal editing process should be followed, with the history preserved. Move any unsupported controversial material to the talk page, and work out a reasonable compromise. Dhaluza 00:35, 21 June 2007 (UTC)[reply]
Comment. What exactly would we be keeping? What could be proposed as a neutral topic for an article? If reliable sources can be found to show that Akl's work is notable, any appropriate work could be merged to our article on Hypercomputation. Note that there is already an article called Selim Akl that contains a statement that's clearly POV: Dr. Akl has shown that the notion of universality in computation is false. This statement lacks reliable sources to establish its truth. I still maintain my Delete vote for the present article, and invite discussion as to whether any of this material belongs in Hypercomputation. EdJohnston 01:54, 21 June 2007 (UTC)[reply]
You would keep what is verifiable and place it in context, even if that leaves a stub. The theory was published, and even if disproved, it is still notable as a rejected theory.Dhaluza 10:07, 21 June 2007 (UTC)[reply]
Comment. One of the consequences of "publish or perish" is that almost any assistant professor has managed to formulate several ideas on which they have written a series of peer-reviewed publications; see also Least publishable unit. Peer review examines whether the work is original, of interest to researchers in the area, and does not contain serious mistakes. Acceptance does not mean it is notable in any usual sense of the word. That comes only when others pick up the idea and start using it (and not just dutifully refer to it in a "related work" section). Should we lower the threshold to publication by the inventor of an idea in a few peer-reviewed publications being enough to consider the topic sufficiently notable to become encyclopedic, there is no end to the uninteresting stuff we can expect.  --LambiamTalk 07:17, 21 June 2007 (UTC)[reply]
You raise a valid point about "publish or perish", however you use this to justify a novel interpretation of WP:RS. There is no requirement there for checking that sources have themselves been cited. Unless the journals publishing this work are known for publishing quackery, we need to defer to their judgment, and not substitute our own. They do more than just check for typos; they decide whether to publish or not to protect their reputation. While this is not an absolutely effective filter, it does eliminate most of the usual wacky stuff. This AfD is taking the WP "quackery patrol" into new ground, arguing the relative merits of the theory itself, and dismissing what appear to be RS. That is not our job. Dhaluza 10:07, 21 June 2007 (UTC)[reply]
He does not need to argue verifiability; merely notability. Papers that are not cited are the equivalent of articles that end up on a cuttingroom floor; there's no evidence they're of interest to anybody. Septentrionalis PMAnderson 22:21, 21 June 2007 (UTC)[reply]
Is that your personal opinion, or does it reflect a consensus opinion shared by the WP community? It certainly seems to go well beyond the policy guidance at WP:V and WP:RS. Dhaluza 00:34, 22 June 2007 (UTC)[reply]
Comment. Hello, about the comment on "publish or perish"... once again, I feel as if some people are very quick to judge these sources or these articles without going through a thorough fact check. I do agree with Dhaluza and beleive we have shown that the article is verifiable. I do not believe Dr. Akl falls under the publish or perish mentality. I can see there being more of a concern if Dr. Akl was a recent PhD graduate, but according to his bio he has had his PhD for thirty years and is a Tenured professor at one of Canada's best universities... i don't see this as a publish or perish motive, and feel to acuse this on that basis is stretching it. I am aware, as per Lambiam 's and my discussion on the talk page that we are to forget all credentials in WP, however, if we are talking about something like publish or perish, his credentials should come into question. Also, I do not think that just because an article about the Perspex machine is being considered for deletion that this is a valid reason to throw the article away. Once again if you read the article, James Anderson dedicates an entire section to Dr. Akl's result. He goes one step further and claims to have found the universal machine... however, this does represent valid research. I don't understand how anything that is not publish in anything less than the most prestigious journals would gain any validity in WP, if editors could not agree as to what is considered a "good" or "bad" journal. The scientific community accepts peer reviewed journals and even though some may be more prestigious and better edited than others, they must all start somewhere. I don't think we can be the judge as to what counts as "Good" peer reviewed sources and bad ones. It seems far too subjective. You either create a list of journals that WP accepts as valid or you must accept them all as valid. Thank you ewakened 00:06, 22 June 2007 (UTC)[reply]
While I basically agree with most of what you said, I think we can give relative weight to different journals, and discount those that have been shown to publish unreliable material. In fact WP:FRINGE deals with this in the "primary witness to notability" criterion. So, we would not use references from a journal less reliable than WP for notability, because that could make WP the primary witness. But in this case, no one seems to question the reliability of the journals, the argument seems to set a very high bar, by requiring that the references are also used as references. This is novel, and in my opinion excessive. Dhaluza 00:29, 22 June 2007 (UTC)[reply]
I think there is a misunderstanding here. The requirement is that others than the inventor pick up on the topic: his or her idea/invention/.... If that is done in academic publications, the authors of these publications will cite their sources, which possibly are the original publications, but that is besides the point. An example. I can easily find many sources referring to Riemann's xi function, even without going back to the original publication by Riemann. Therefore his xi function is notable. The same does not hold for his psi function defined in the same publication. Although Riemann's publication Ueber die Anzahl der Primzahlen unter einer gegebenen Grösse is well cited, the psi function he defines therein is rarely mentioned. It is not notable. If the mere fact of an idea being used by its inventor, who happens to be the author of a publication mentioning it, is enough to establish notability, then we are definitely setting the bar too low.  --LambiamTalk 09:00, 22 June 2007 (UTC)[reply]
I agree that sources have been cited are cited can be considered more reliable, and given more weight. But, the bar for inclusion has been set much lower in WP policy, and while the arguments to raise it may have merit, they do not represent consensus. The place to try to get consensus for new policy is in the related policy forums, not AfD. Dhaluza 09:09, 22 June 2007 (UTC)[reply]
I don't understand why you keep bringing up reliability, since that is not at issue here. The place where we try to get consensus on the question whether the topic is sufficiently notable to warrant inclusion in our encyclopedia is right here, in this discussion. Also, I don't know which of the many Wikipedia policies you are referring to when you write "the bar for inclusion has been set much lower in WP policy". I am not trying to set any new policy, but am just applying what I consider to be common sense. There are some guidelines for notability – which is not the same as policy – but they are not cast in stone, and are to be applied with common sense.  --LambiamTalk 10:06, 22 June 2007 (UTC)[reply]
OK, so then you agree that the sources are reliable per WP:RS (and therefore at least some of the content is verifiable per WP:V). Where your interpretation goes well beyond policy is by rejecting these sources on some other grounds, then saying it does not have Notability (which is based on sourcing). You need to cite some consensus based policy or guideline supported argument for deleting this, otherwise what you call "common sense" seems to me to be WP:IDONTLIKEIT. Dhaluza 10:23, 22 June 2007 (UTC)[reply]
Could you be more specific in identifying the relevant policy than saying "well beyond policy"? I have no clue what the policy is or policies are that you are referring to. I think the material covered in the article is non-notable. If you feel that it is notable and therefore should be kept, please argue why you think that it is notable, instead of waving some unspecified policy in my face.  --LambiamTalk 13:11, 22 June 2007 (UTC)[reply]
P.S. Of course the sources are reliable, in the sense of WP:V. They are primary sources. What source could be more reliable for verifying that Dr. Akl claims such-and-such than a publication by Dr. Akl in which he claims such-and-such? 14:08, 22 June 2007 (UTC)
This statemet is contrary to the WP definition of primary and secondary sources at WP:PSTS. The author's notes are primary sources, but his scholarly publications are secondary sources. This argument is conflating the author with his theory. The author is a primary reference in relation to his biography, but not his theory. Again, these new, novel, and evolving deletion arguments all appear to be contorting policy to support what at bottom is WP:IDONTLIKEIT. Dhaluza 01:12, 23 June 2007 (UTC)[reply]
That is not my understanding of WP:PSTS for mathematical publications. Primary sources are the original published research papers describing a new concept, theorem, definition, etc. Secondary sources are survey papers, textbooks, or other research papers that describe the same material but do not make any claim of novelty for it. Research papers that cite the original published work may be secondary sources, or may not, depending on how trivial the citation is. Author's notes are generally not available and therefore generally not any kind of source at all. —David Eppstein 05:22, 23 June 2007 (UTC)[reply]
WP terms have definitions that may differ from common or specialist usage, but we must use the WP definition to interpret WP policy. WP:PSTS does not address mathematics specifically (the term "math" does not even appear), but it does address science in general, linking to lab notebook under primary sources. Mathematics research is a little different, but not so different that we need a completely different definition. Primary sources on WP are the set of initial observations, and secondary sources are the subsequent conclusions. Wikipedia is a tertiary source that reports on what is published in RS. There is no need for WP to set a higher standard for inclusion by only considering secondary sources that have been cited previously. The authors of the consensus-based policy documents could have used this standard for inclusion, but did not. Dhaluza 10:23, 23 June 2007 (UTC)[reply]
Although not fully accepted as consensus, I think WP:SCI is far more relevant to this discussion than the quotes you are mining from WP:PSTS, especially since lab notebooks have little relevance to this type of research. —David Eppstein 10:36, 23 June 2007 (UTC)[reply]
There should be a secondary source (preferrably by someone other than Dr. Aki) that the material is WP:Notable, or that it means what the editors here say it does, or we've violated WP:SYN. — Arthur Rubin | (talk) 19:25, 22 June 2007 (UTC)[reply]
This is not in keeping with the letter or the spirit of that guideline either. WP:SYN deals with synthesizing material in articles on WP, not synthesizing ideas as scholarly research, publishing those ideas in WP:RS and then citing them on WP. This is specifically permitted in the next paragraph of that guideline WP:COS: "Citing oneself: This policy does not prohibit editors with specialist knowledge from adding their knowledge to Wikipedia, but it does prohibit them from drawing on their personal knowledge without citing their sources. If an editor has published the results of his or her research in a reliable publication, then s/he may cite that source while writing in the third person and complying with our NPOV policy." Dhaluza 00:46, 23 June 2007 (UTC)[reply]
  • Keep and really clean-up There appears to be a larger body of published work on this topic than is represented in the article currently. A cursory Google search turned up published works by several different authors including Oron Shagrir, Gualtiero Piccinini, Toby Ord, Tien D. Kieu, Cotogno, and Svozil in a wide variety of journals and books. Research appears to be progressing at several locations around the world, from Australia to Israel to the U.K. to Canada to St. Louis. This leads one to believe that the ideas presented appear to be widespread in the academic community, and notable. The article really needs to be cleaned up and better sourced, but that merits stubbing the article, not deletion. SqlPac 00:24, 22 June 2007 (UTC)[reply]
Question. Can you give a pointer to publications by any of these authors that address the topic addressed in the present article (non-universality in computation, in the particular sense as defined by S. G. Akl?)? This article is not about the research of the authors you mention, even if there are common aspects. For most of the topics addressed by the research of these authors we already do have articles that deal with it, such as our articles on Hypercomputation and Quantum computing. I am not necessarily against a different article with a different content on a different topic with probably a different title; my concerns are about the article under discussion here.  --LambiamTalk 06:36, 22 June 2007 (UTC)[reply]
  • Clean up. There's certainly a topic here - but it presents itself in a way that verges close to "zany new theory" in its apparent contradiction of established ideas that in fact stems from inconsistent definitions. It needs a lot of careful clarification and references from major journals. Dcoetzee 01:20, 22 June 2007 (UTC)[reply]
  • Keep I question the motivation of the nominator. It seems that he was not able to develop consensus for his position at the talk page so he nominated this for AfD. I can't say that this is inherently prohibited or wrong, but the move seems premature. Perhaps this should go back to the talk page for an attempted resolution and the AfD should be postponed until we can evaluate the progress. --Kevin Murray 03:21, 22 June 2007 (UTC)[reply]
    Please assume good faith. Indeed, the author and I were unable to establish consensus that the topic is notable. However, the reason for my nomination is not the unremarkable fact that the author did not agree with me that the topic is not notable, but rather that I believe it is not notable.  --LambiamTalk 06:13, 22 June 2007 (UTC)[reply]
  • AfD is not based on consensus, but on the strength of people's arguments - the closing admin should be like a judge, weighing all arguments and deciding which is the best option. So I don't see a problem.—greenrd 09:33, 23 June 2007 (UTC)[reply]
  • Merge and redirect to Selim Akl. This should not be presented as an accepted theory until there is some evidence that it is actually accepted; as it stands, the concept covered in this article is purely the product of Akl's work, and is therefore only appropriate in Wikipedia as part of our biographical coverage of Akl. However, the author has made a solid argument that this does comply with WP:V and other core policies; therefore it's reasonable that we include it somewhere. -- Visviva 10:16, 22 June 2007 (UTC)[reply]
    • Comment moving hotly disputed content to a WP:BLP page is probably ill advised. If the content were stable, a merge might be an option. Having a separate article does not mean it is an accepted theory, and the article can and should clearly lay that out. Dhaluza 10:38, 22 June 2007 (UTC)[reply]
      • Well, if the material is basically biographical -- that is, pertinent only to our coverage of the person, rather than the field -- then BLP applies regardless. -- Visviva 10:50, 22 June 2007 (UTC)[reply]
        • I don't think the theory was intended to be an autobiography, so we shouldn't cast it that way. Also some of the refs were co-authored, so it is not strictly related to one individual. Dhaluza 11:30, 22 June 2007 (UTC)[reply]
        • Dhaluza is right that the article was not meant to be autobiographical. Dr. Akl might have been the first to put himself out there with these ideas, but he is certainly not the only one, so to call this work autobiographical would be taking away from others as mentioned by Dhaluza. ewakened 12:36, 22 June 2007 (UTC)[reply]
          • I don't dispute that the article is not intended to be biographical. But the article as it stands is not about the general theory (if there is one), but about how that theory has been put forth by one person. That makes this effectively part of our biographical coverage of that person, whether we like it or not. -- Visviva 15:40, 22 June 2007 (UTC)[reply]
  • Comment. Even if the concept were notable, the name is only used by Selim Akl, so WP:NEO comes to mind a reason to delete. — Arthur Rubin | (talk) 13:29, 22 June 2007 (UTC)[reply]
    • This argument also is a novel interpretation of that guideline. And,like the others, it turns it on its head by arguing that something specifically permitted in the text is not OK to do. If you read all the way to the end, you'll find "To paraphrase Wikipedia:No original research: If you have research to support the inclusion of a term in the corpus of knowledge that is Wikipedia, the best approach is to arrange to have your results published in a peer-reviewed journal or reputable news outlet and then document your work in an appropriately non-partisan manner." Dhaluza 23:35, 22 June 2007 (UTC)[reply]
      Quoting from the same paragraph: "To support the use of (or an article about) a particular term we must cite reliable secondary sources such as books and papers about the term—not books and papers that use the term." Actually, if you do a Google search for "non-universality in computation", all you get is links to this article and the Selim Akl article. With the search "universality in computation" you additionally get references to Turing, as in "Alan Turing defined the abstract Turing Machine which clarified the concepts of computability and universality in computation", so then it is used in a different sense than the particular meaning of Akl. That is even stronger with the search term "universal computation" – the term is generally used in the meaning of "capable of computing any computable function". By the way, my espresso machine can compute espresso given ground coffee and water as input, something no Turing Machine can do. However, I'm not going to offer this as a counterexample to Church's Thesis.  --LambiamTalk 10:07, 23 June 2007 (UTC)[reply]
      I did not see this as an article about a neologism as a word, just one that uses a neologism as its title, which all articles about something new or different must do. But we don't delete articles with bad titles, we rename them. In fact, you may have stumbled on a reasonable compromise to the content dispute. Why not just rename this to Universality in computation and describe the work by both Akl et al., and Turing? Dhaluza 10:35, 23 June 2007 (UTC)[reply]
      Why not? Because this work by Akl, as well as his particular use of the term, is non-notable, that's why.  --LambiamTalk 11:01, 23 June 2007 (UTC)[reply]
      We already have an excellent treatment of the Turing machine and Turing's other work here at WP. The article you propose would invite an undue comparison of Turing and Akl i.e. presenting their theories as equally established, which is certainly not the case. Perhaps Akl's work could be mentioned in a couple of sentences in one of the article's that already speaks to these issues.--Cronholm144 11:39, 23 June 2007 (UTC)[reply]
      Yes, I saw that and the one at Universal Turing machine, Turing machine examples, Post-Turing machine, Turing completeness, and utm theorem. So with all this coverage, a comparison with a competing theory would not be undue, it would provide balance. The two theories do not need to be presented as equally established, obviously one has been around longer and gained greater acceptance. But we do not censor or dismiss minority opinion or alternative theories, we present them in context. The Non-universality theory has cleared the bar in terms of notability by being published in reliable scholarly sources by a recognized academic author. Whether it is true or not is not for anonymous editors to decide on WP. Dhaluza 13:54, 23 June 2007 (UTC)[reply]
      If we presented the two theories with the weight that notability would dictate that we would have to write an article in a 300:1 (Turing et all) to (Akl "et all") ratio.--Cronholm144 14:45, 23 June 2007 (UTC)[reply]
  • Delete. I've refrained from commenting thus far, as it seemed to be that Akl's publications have established a marginal grounds for keep. Nonetheless, as this discussion has progressed, it's become clearer to me that the people arguing "delete" have a better grasp of Wikipedia policy and how Akl's work here pertains. Akl's challenge to the notion of universality in computation is one of many by others that have been made over the years, and like the others, shows no sign of making a non=negligible impact on the field. Dhaluza's proposal to promote Akl's work in order to "give balance" is ridiculous. Why should we promote Akl's work over all the others that have also raised challenges of dubious merit to Turing's work? There is no reason to regard Akl's work of worthy of even mention in Wikipedia. There are a few peer-reviewed publications, sure, which incidentally appear in journals with Akl as an editor. The others do not appear peer-reviewed. As a big section of WP:RS states, extraordinary claims require extraordinary evidence. Just citing a small number of peer-reviewed publications by one researcher and his collaborator does not qualify as an "extraordinary" evidence. --Chan-Ho (Talk) 23:28, 23 June 2007 (UTC)[reply]
  • Keep. The article has been cleaned up. Now it makes some sense and maintains NPV. Jmath666 15:45, 24 June 2007 (UTC)[reply]
    • Comment. It makes sense, now, although no sources have been provided to indicate that anyone other than him (and possibly Carl Hewitt) thinks it notable. No change in my "delete" !vote. — Arthur Rubin | (talk)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Articles_for_deletion/Non-universality_in_computation&oldid=140323181"