Wikipedia:Articles for deletion/Direct logic Source: en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion/Direct_logic
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.
This page makes dubious claims. It changes in response to talk page remarks, but sources are not provided, or provided sources are irrelevant -- certainly nothing in a mathematical publication. I'm afraid it fails the smell test. -Dan (User:Fool) 15:21, 8 June 2006 (UTC)~[reply]
Weak keep There's a long, sad story here. The main author of the article, Anonymouser (talk·contribs), is believed to be a sockpuppet of Carl Hewitt, who is referenced in the article. Hewitt was a major figure in the logic-based end of artificial intelligence back in the 1970s. This is widely believed to be a dead end today, and Hewitt is one of the few people still trying to make it work. The general idea here is to develop "weak logics" which have more tolerance for ambiguity and contradiction than classic logic, which is very brittle. John McCarthy developed "circumscription" as an early effort in this direction, and there have been other variations on this theme. None of them are really all that useful. Bayesian inference seems to work better. What we're seeing here is something that, for now, seems to be a theoretical dead end. That happens in mathematics. The article does cite published papers, though, and I'm inclined to keep it in. --John Nagle18:09, 8 June 2006 (UTC)[reply]
Delete. The logic-based approach to AI has been revived the last few years in the forms of agents (similar to Hewitt's actors) and they may indeed have a future (my CS department is gambling on it) but this is of little significance. The article is based on two conference papers by Hewitt which do not seem to have been published in a peer-reviewed journal making this original research. (And even if they were published in a peer-reviewed journal there should be at least few other authors writing about or refering to Hewitt's direct logic to make this verifialbe and notable enough for inclusion.) —Ruud04:43, 11 June 2006 (UTC)[reply]
Since the publication date is March 2006, we may have to wait awhile for other publications to refer to it. —The preceding unsigned comment was added by 64.251.124.7 (talk • contribs) 17:19, June 11, 2006 (UTC)
Comment. The author cites his own "published" (but not generally available) papers. Even if they are generally available papers in a peer-reviewed journal, they still violate his RfA, as noted above. See Hewitt (2006a) and (2006b) — both seem to be invited, but not peer-reviewed, papers. As (2006a) is referenced in the lead, the entire article is suspect. (One of my published papers was invited at a conference, but not peer-reviewed — I know better than to attempt to reference it in Wikipedia.) — Arthur Rubin | (talk)18:33, 8 June 2006 (UTC)[reply]
Actually both papers seem to have been peer reviewed although the COIN paper more rigorously than the AAAI paper. Also invited papers count as publications for the purposes of reference for the Wikipedia. In science there are many examples of invited papers that published important results. —The preceding unsigned comment was added by 72.254.25.74 (talk •
There are many examples of invited papers which are not peer-reviewed. We shouldn't accept those until they are at least commmented on in a peer-reviewed paper, even if important. — Arthur Rubin | (talk)22:37, 8 June 2006 (UTC)[reply]
Peer review is especially important where dubious claims are involved. -Dan 02:02, 9 June 2006 (UTC)
Keep The grounds originally cited for deletion are not valid. The article reports on a publication by a noted academic. The results of the publication have been presented in seminars at Stanford and SRI. So it is difficult to argue that it is not notable. The field is divided into camps that radically disagree with each other. Consequently the ultimate value of the research is a matter of great controversy. However, the problem addressed in the motivation section of the article is of the highest importance in computer science: What is the field going to do about the pervasive contradictions among documentation and code of large software systems (think Windows!)? Also there do not appear to be any glaring errors in the article. — Preceding unsigned comment added by 72.254.59.254 (talk • contribs)
As in my comment above, Carl Hewitt may be "a noted academic", but papers (2006a) and (2006b) are from conference proceedings, in which it appears he was invited, but the papers were not peer-reviewed. (And it's still fringe mathematics, rather than dead-end mathematics, even with the 20-year history. CS people, when trying to do mathematics, sometimes get confused.) If anything related to those papers is excised from the article, and there's still something there, I'd make it a reluctant keep. Those papers are invalid as primary or secondary sources, except as an indication of the opinions of the author. — Arthur Rubin | (talk)18:33, 8 June 2006 (UTC)[reply]
Comment To add to that, as I stated on the article's talk page, (2006b) is not even a relevant reference. As for the possibility raised about the editor's identity, well, I suppose it explains where the editor gets his info from... but then it means Hewitt has taken to refer to himself in the third person! Not likely is it? -Dan 19:08, 8 June 2006 (UTC)
Could the IP 72.254.59.254 (from South Jordan, Utah) please present evidence that it is not another sock puppet ofCarl Hewitt? Its edits on Talk:Direct logic do give that impression.-- ZZ09:29, 9 June 2006 (UTC)[reply]
Paraconsistent logics generally are among those "sciences" (like fuzzy "logic") which I consider to be scams. We should not dignify any of them with coverage which takes them seriously, i.e. at face value. If we mention them at all, it should be to expose them as scams. JRSpriggs06:02, 9 June 2006 (UTC)[reply]
Delete. It is neither peer reviewed research nor are the references to published research shown to be relevant (they are POV actually). Instead, direct logic evolves while we watch. -- ZZ09:14, 9 June 2006 (UTC)[reply]
Keep The Wikipedia article is about the views of the publication referenced in the article. If other people have different views they should put them in the article. — Preceding unsigned comment added by 72.254.20.137 (talk • contribs)
More sock puppetry from South Jordan, Utah, repeating the claim that there was a publication and ignoring that it did not receive a peer review. -- ZZ12:08, 9 June 2006 (UTC)[reply]
South Jordan, Utah? I am at a hotspot in San Francisco talking with friends about this. How do you know that the publication was not peer reviewed?
Given that it is not generally available, how do we know it was? How do we know anything at all about Direct logic? If we are supposed to take your word on all of this, are we really being unreasonable to ask you to identify yourself? -Dan 18:02, 9 June 2006 (UTC)— Preceding unsigned comment added by 192.75.48.150 (talk • contribs)
Keep Hewitt gave a seminar on this at Stanford. It seemed OK. —The preceding unsigned comment was added by 171.64.71.47 (talk • contribs) 22:51, June 9, 2006 (UTC)
Weak Keep. Although it may be original research, the research appears highly validated and relevant. It is on these grounds that I am reluctantly vote to Keep it. -- Evanx(tag?) 20:32, 11 June 2006 (UTC)[reply]
So, it is original research - which has no place in a Wikipedia. As for your other points, the discussion shows that not even the basics of "Direct Logic" have been established (does it explode or not?). Anyway, if it is so relevant, Hewitt can surely publish it in a peer reviewed journal. Then, he can place it on Wikipedia. -- ZZ10:10, 12 June 2006 (UTC)[reply]
Even if we did accept OR (which Wikipedia doesn't under any circumstance), Direct Logic is not "highly validated". The introduction claims it is paraconsistent, but in the caveats section we read that it has not been proven that it doesn't explode. —Ruud11:07, 12 June 2006 (UTC)[reply]
And the claim may be very hard to prove. These things are known as conjectures and some have become famous (e.g., Fermat's). Sometimes they take a very long time to prove (or disprove). --2ndMouse13:19, 12 June 2006 (UTC)[reply]
Delete. Since no one else except Hewitt has published about this, I think we can consider it to be non-notable at the moment; eventhough the author is notable. The article can always be re-introduced once more publications by other people have appeared. -- Koffieyahoo01:31, 12 June 2006 (UTC)[reply]
Keep; the verifiability and no-original-research policies are satisfied by previous publications, and notability stems from the preminency of the author. - Liberatore(T) 20:08, 13 June 2006 (UTC)[reply]
While the publications are discussed above and I have nothing to add to that aspect, I must say I don't agree that everything someone notable writes is itself notable. -Dan 03:35, 14 June 2006 (UTC)
Keep the presentation of this stuff at MIT was way cool. —The preceding unsigned comment was added by 18.85.18.103 (talk • contribs) 21:41, June 16, 2006 (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.