Talk:Squashed entanglement

Someone has objected to this page as being a possible vanity page because "almost all references are unpublished preprints." The preprints are in ArXiv. They can be accessed by anyone. They have a time stamp. Thus they are published. Publishing does not mean publishing in paper. I haven't seen you point out anything incorrect in the article or the preprints. Superperro (talk) 23:09, 5 May 2009 (UTC)[reply]

Hi, yes, that was me. Perhaps I should have said "self-published" instead of "unpublished", but for Wikipedia purposes they are the same. Wikipedia's standard of documentation for this type of material involves citations to peer-reviewed journals. If some are available, could you fix up the article appropriately so that it does not appear to contain original research? I left a note at WT:PHY asking for someone else to look at the article but got no response, so I prodded it. Thanks. (Note, I left a further request there.[1]) 67.122.209.126 (talk) 06:29, 6 May 2009 (UTC)[reply]

If you had taken just a few seconds to follow the arxiv links, you would have found out that many of the referenced papers have indeed been published in peer reviewed journals. I have use Citation Bot to complete these references. The only references that appear to be suspect are the ones by Robert R. Tucci. None of these have been published, in fact I have only found one of his 32 papers on arxiv published. So there might be an issue with undue weight given to his work, which would need further inquiry. (which I don't have the inclination to do for the moment.) (TimothyRias (talk) 09:07, 6 May 2009 (UTC))[reply]

Well, it doesn't appear that Tucci believes in publishing in peer reviewed journals, but at least ArXiv puts an *indelible* time stamp on all papers. Of course, wikipedia (or many other great internet resources) are not peer reviewed either, but erasing them for that reason would be a great loss for humanity :) You can easily check from the article's citations that the assertions about Tucci's work are accurate. You can check, for example, that reference Chr03 cites within it two of Tucci's eprints as their reference 26 Superperro (talk) 10:44, 6 May 2009 (UTC)[reply]

Right, Wikipedia is a tertiary source, which means that its reliability derives from citing secondary sources like refereed journal articles. It "outsources" the refereeing process to the journals instead of trying to evaluate original research itself. It's fine that Tucci doesn't believe in publishing in refereed journals, but a consequence is that Wikipedia in principle is not supposed to cite Tucci. I'm not as big a stickler for this particular policy as some editors are, but the reasons for it are sound and I'm not persuaded that this article deserves an exception. Anyway, I'll wait and see what others have to say. I see that Timothy Rias has improved the references somewhat, which is promising. 67.122.209.126 (talk) 14:39, 6 May 2009 (UTC)[reply]

Could you please point out where wikipedia states the rule that you allude to. Forgive me, but it sounds like an odd rule indeed. Even Peered reviewed Physics journals (like Physical Review) allow their authors to cite unpublished ArXiv references. How about computer manuals and software, should wikipedians abstain from citing those too, because they are not peer reviewed? And the existential question, since wikipedia is not peer reviewed, is wikipedia justified in citing itself? :) And how about the article about Perelman's proof of the Poincare conjecture. Perelman's papers can only be found in ArXiv. Should we erase those too? Another example, in quantum computing this time, is the paper by Coppersmith where he invented the Quantum Fourier Transform decomposition. That was never published in a peer reviewed journal but is often quoted. See, for example, this American Mathematical Society (AMS) article: http://www.ams.org/featurecolumn/archive/quantum-one.html Superperro (talk) 17:46, 6 May 2009 (UTC)[reply]

Superperro,

The relevant Wikipedia rules to your first question are: Wikipedia:Verifiability, Wikipedia:Reliable sources, Wikipedia:No original research, and maybe some others. Please read these before asking further questions of that nature. Wikipedia:Conflict of interest and Wikipedia:Tendentious editing may also be of relevance.
The rules for peer reviewed journals and the rules for Wikipedia are completely different, so it should be no surprise that some stuff acceptable in one is unacceptable in the other. To state the obvious, journals exist for the purpose of publishing novel results a/k/a original research, which is strictly forbidden on Wikipedia. The scientific community trusts the contents of the journal articles because of the quality control supplied by the refereeing process, where the editors and referees are recognized experts in the relevant field. That process of course occasionally makes errors, but it's far more rigorous than something like Wikipedia (edited by millions of anonymous people with possibly no qualifications at all) can ever hope to ascribe to itself. So Wikipedia chooses to trust the judgement of journal editors to assess the quality of research results rather than trusting its own editors' judgement. If the journal editors haven't accepted a result yet, Wikipedia doesn't accept it either.
No, Wikipedia is not allowed to cite itself as sourcing for factual info, since as an open wiki, its contents are considered self-published (see Wikipedia:Verifiability). Of course Wikipedia articles extensively cross-link to one another as an aid to browsing, but that's different from self-citing as a source.
The sourcing expectations are different for different subject areas of Wikipedia. E.g. for articles about computers, trade journals and manuals can be ok as sourcing. For articles about recent political events, newspapers can be ok. For celebrity romance, even gossip mags might qualify. For specialized research topics in academic areas like quantum physics, we're looking for refereed journal articles.
The comparison between Tucci and Perelman will become interesting when the ICM tries to award Tucci a Fields medal like it did Perelman. Until then, not so much.
If Coppersmith's eprint about the Quantum Fourier Transform is quoted in a refereed AMS journal, then the quotation and its cite are part of the refereed contents and we can use them if we cite the AMS journal. That doesn't automatically make everything in the QFT eprint usable. For example, if Coppersmith's eprint proves five theorems and the AMS article quotes two of them, we can use only those two. We would say something like "Joe Schmoe (give cite of AMS article including issue and page number) cites a result by Coppersmith which says...".

Hope this helps. 67.122.209.126 (talk) 08:35, 7 May 2009 (UTC)[reply]

Thanks for the reference to the rules. I wasn't aware of them, although, as primary author of this article, I don't believe I have violated them.

If your point is that the Tucci references should be omitted, I don't see how this can be done with any fairness. In fact, omitting the Tucci references and leaving the rest, would be, IMHO, pushing a very insidious agenda of hiding the truth about the origins of this idea. I also want to repeat 2 points made earlier (1)Tucci's work has been cited in subsequent peer reviewed journals. I already gave an example: Chr03 (published in a peer reviewed journal) cites within it two of Tucci's eprints as their reference 26. I can find other examples if you want. (2)Tucci's eprints are in ArXiv, which is a highly verifiable source, since it gives articles an indelible time stamp and makes articles available for perpetuity.

As primary author of the article, I also vouch for its accuracy, I have a PhD in Physics, and have studied carefully the subject of this article. If you can point out an error in the article, even about my assertions about Tucci's work, then I will gladly change it to accommodate you.Superperro (talk) 19:08, 7 May 2009 (UTC)[reply]

I concur with Timothy Rias's comment (that I just noticed today). He has added good citations that indicate that the article topic is well enough documented that deleting the article isn't called for. But he (and I) see the possibility that Tucci's work is over-represented in it (see: Wikipedia:Undue weight). Re the Tucci references, it's like your Coppersmith example. Two possibilities:

a) Chr03 states and cites Tucci's theorem XYZ, and the Wikipedia article also cites this theorem to Tucci. In this case, don't remove the Tucci cite, but supplement it by specifying exactly where Chr03 referenced it. E.g., change "theorem XYZ (Tucci02) to "theorem XYZ (Tucci02, cited by Chr03 page 17)". Page 17 should be the page in Chr03 that actually mentions theorem XYZ, not the place in the reference section at the end where the eprint url appears.

b) The article mentions and cites Tucci's theorem ABC, but that theorem is not mentioned by Chr03 or any other refereed paper. In this case, leaving in the statement ABC but omitting the citation would indeed be unfair. The cure is to remove both the statement and the citation. A strict adherence to policy would require this, but I'll leave it to others to decide whether strict adherence is called for here. Certainly as far as Wikipedia sourcing problems go, this article's are pretty mild compared some of the stuff we get.

I see you are very interested in writing up Tucci's research and it has occured to me that the two of you might, um, know each other. Your best bet may be to persuade Tucci to submit his work to a journal, so that the referees can decide on its accuracy and relevance. If they print it, that solves the whole problem about citing it here. Note that "verifiability" on Wikipedia means much more than just being able to check that something exists. For explanation, please see the verifiability policy page that I linked to earlier.

To Timothy Rias: Thanks very much for adding those cites. I remember I did find a few of them in Google Scholar, but you located a bunch that I missed. 67.122.209.126 (talk) 11:54, 9 May 2009 (UTC)[reply]