Talk:arXiv

Wiki Education Foundation-supported course assignment

This article was the subject of a Wiki Education Foundation-supported course assignment, between 24 August 2020 and 4 December 2020. Further details are available on the course page. Student editor(s): Paxrei.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 14:36, 16 January 2022 (UTC)

'The ArXiv'? or just 'ArXiv'

Amongst physicists the site is always referred to as 'the arXiv', which parts of this article reflects. Other parts of the article miss out the 'the'. It should be consistent across the entire article, so should it be 'the arXiv' or just 'arXive' throughout? Apparently this is a point of contention as whether you call include the definite article or not seems to be field-dependent. (Mathematics and Engineering friends of mine just call it 'arXiv'. Yuck.) — Preceding unsigned comment added by 163.1.19.1 (talk) 09:16, 26 April 2017 (UTC)

The website clearly refers to itself as simply 'arXiv':

"arXiv is an e-print service in the fields of physics, mathematics, computer science, quantitative biology, quantitative finance and statistics. Submissions to arXiv should conform to Cornell University academic standards. arXiv is owned and operated by Cornell University, a private not-for-profit educational institution. arXiv is funded by Cornell University Library, the Simons Foundation and by the member institutions."

This should be reflected in the Wikipedia page by the removal of all preceding "the"'s. — Preceding unsigned comment added by Rb09 (talk • contribs) 09:23, 26 April 2017 (UTC)

References

What's in a Name

anyonhe know how the name arXiv came about? I'm guessing its a pun on 'archive' with X coming from the old 'xxx' name? But why no 'e' on the end?

Their FAQ says "no reason" for both "why arXiv.org?" and "why xxx.lanl.gov?". HTH. -- ALoan (Talk) 12:02, 22 September 2006 (UTC)

In Russian language letter "Х" always gives sound "kh", like "X" in Donald Knuth's "TeX" program. My version: Any native Russian tends to read "arxiv" as "arkhiv" (not an "arksiv") which sounds exactly like "архив" and means in translating exactly "archive". In Russian language there in not anything like the speakless "e" letter. Another example of such weird writing: "KAPABAH" - every Russian knower will see here first of all absolutely exactly written "caravan" word (Караван/караван/КАРАВАН). — Preceding unsigned comment added by 46.201.167.217 (talk) 16:41, 16 June 2016 (UTC)

License

Question: can arXiv.org be described as open content? What is the licence or other conditions that arXiv materials are released under?

I do not know about the licencing, but this question does rather not occur in most uses of the medium, where people want to learn information, but not modify or redistribute it. I think that (i) authors will generally agree to redistribution of their works, and (ii) authors reserve the right to remove their work from the public domain (which could for example be needed when an article appeared in a journal that forces the author to do this [sadly this already occured -- John Baez discusses such issues on his homepage], in which case the author will hopefully no longer submit to or referee for this journal). ArXiv certainly is part of the Open access movement -- I will add a link. --Markus Krötzsch 00:54, 6 Jul 2004 (UTC)

There are several licenses submitting authors can choose from. The one which releases the fewest rights to arXiv (IE the minimum to put something on there) grants non-exclusive and irrevocable license to distribute the article, linked below.

http://arxiv.org/licenses/nonexclusive-distrib/1.0/license.html

You can also select public domain or a couple creative commons licenses. One frequent use I didn't see mentioned is for citations of new works - since the review/publication cycle can be quite long in some cases, you can have a paper on arXiv many months before it's actually published. For people submitting papers to journals in the meantime, arXiv gives a way to reference work that's been submitted/approved but not published yet.66.57.254.204 (talk) 05:08, 17 September 2010 (UTC)

XXX

The following paragraph currently in the article is wrong:

arXiv.org was formerly known as xxx.lanl.gov but it changed its name when it was found that censorware programs were blocking access to it from various sites, under the impression that the three letters 'X' in its name implied that it was a pornographic site. The idea of XXX was that it was one better than WWW in every way.

XXX was chosen by Paul Ginsparg as a joke to try to get the site and it's emails blocked. See http://arxiv.org/help/faq/skullfaq for more info.

Also, the entire article is slanted towards mathematics, and there's no mention of the nlin category (non-linear sciences).

arXiv surpassed 300,000 papers in November 2004.

I'd make the changes myself, but I work for the arXiv, and thought that it could be a conflict of interest. So I'll let someone else in the community edit the page as a buffer between my POV and the the NPOV of the article. ktheory 18:58, 11 Mar 2005 (UTC)

I updated the article count and added a mention of the nlin category. The FAQ (http://arxiv.org/help/faq/skullfaq) you are pointing to is entirely unconvincing. I can't help with the slant towards maths, as I am a mathematician myself, but feel free to edit (I think the conflict of interest is no problem once you have declared it). -- Jitse Niesen 17:51, 30 Mar 2005 (UTC)

I seem to recall Stevan Harnad saying it was because XXX was one better than WWW, when I questioned the wisdom of xxx as a domain name when were setting up a mirror for it back in the 90s. Elseware (talk) 08:12, 3 April 2019 (UTC)

For posterity: the FAQ page is dead, Wayback has it at http://web.archive.org/web/20080517143155/http://www.arxiv.org/help/faq/skullfaq . The relevant part is

"For a time I was not receiving any mailing list missives. Then I discovered the reason -- I had a spam filter set up to catch "xxx" (because that's a typical earmark of porno-spam). Why did you chose the name...?

Because the majority of our mailings are porno-spam, and we prefer to make it easy for discriminating filters to block us. (Anti-nuclear protestors are also welcome to block "lanl", and anti-government activists are welcome to block "gov" -- go for the triple block.) Thank you for your interest."

I agree with Jitse Niesen that this is just a joke. Personally I'm happy to take Elseware's word for it regarding the www/xxx story. Patallurgist (talk) 21:04, 28 October 2022 (UTC)

Incorrect Title

Why not just move the title to "arXiv.org e-print archive" I dont think that theres any limitation.

MediaWiki capitalises the first letter of article names. That is why iPod is at IPod. The reason for this is so that there is no need to worry about the case of something when linking it ie you can talk about "how fast the car went" and also that "Cars can go fast". Both links go to the same place. 132.181.7.1 00:41, 9 Jun 2005 (UTC)

Template?

Is there a template for arxiv referencing? --Staecker 15:30, 20 October 2005 (UTC)

The template {{arXiv}} can be used inside a larger reference. For example {{arXiv|hep-th/0203101}} creates arXiv:hep-th/0203101. This can be used in the ID field of templates such as {{Journal reference}}. -- Fropuff 02:46, 19 January 2006 (UTC)

Perelman and Arenstorf

Hi, so what did Perelman and Arenstorf do in 2002 and 2004? (sorry - you've got a physicist here). Did they keep doing it afterwards, or were they one-off events? Just asking because the passage in the article read oddly. I've changed it somewhat, but someone who kows had better check I haven't stuffed up the meaning. More detailed explanation wouldn't go awry, either ;-) Deuar 16:30, 8 March 2006 (UTC)

Why this strange name?

Why does this article have such a strange name? JSTOR is in JSTOR, so why not put this in ArXiv or ArXiv.org? Loom91 11:27, 22 March 2006 (UTC)

It is strange and I have no answer except to say it should be moved to arXiv. There are already a number of redirects from that page and I suspect the only reason there is a large number of links to arXiv.org e-print archive is because people have been fixing the redirects. --C S (Talk) 19:33, 22 March 2006 (UTC)

Children's Internet Protection Act

"It is sometimes claimed that some censorware programs were preventing some users from accessing it at its previous address, xxx.lanl.gov, under the impression that the XXX in its name implied that it was a pornographic site; however, this is believed to be untrue: private internet service providers would not be financially viable if they blocked access to such sites."

This is true for residential ISPs. However, arXiv contains scholarly works that are intended in part to be useful to students accessing the web through a school ISP. In the United States and possibly other jurisdictions, internal ISPs run by the IT departments of public schools (both K-12 and university) and public libraries are required by law to use censorware on student-accessible machines, or they will lose funding. (See Children's Internet Protection Act and foreign counterparts.) Here, censorware increases financial viability. As for private schools, socially conservative parents are likely to send their children to schools that take steps to protect their children from pornography. --Damian Yerrick (☎) 22:04, 7 April 2006 (UTC)

is it true that most papers on the arXiv get published elsewhere?

That little bit of text (currently "some small fraction of work remaining purely as e-prints and not published in peer-reviewed journals") gets quite a bit of editing attention. Does anyone have any evidence for this statement? My impression would be that to the contrary quite a lot of stuff put on the arXiv, perhaps even a majority, never gets published in a peer-reviewed journal. Does anyone have any numbers for this? Dmharvey 12:29, 20 April 2006 (UTC)

Sigh, for now, I'm going to change it to say "some" instead of "some small fraction", "some minority", etc. It's demonstrably true and verifiable that the vague "some" is correct. I don't believe there is any data from a verifiable source (in the sense of WP:Verifiability) that supports the other statements.

As for your impression, I guess it depends on where you look. Deuar and I have discussed this, and our impressions is that in our own specialties (mine is primarily math.GT), about 10% remain unpublished, with the rest getting published within a few years after submission. --C S (Talk) 16:38, 20 April 2006 (UTC)

Ok, I changed it to "some", but to make it more interesting (instead of almost a vacuous statement), I added that this includes even "very influential papers". I didn't find it necessary to cite this as there are, I think, enough examples, but if someone objects, one example is William Thurston and several of his preprints. Some of them were to be published in the Annals of Mathematics, but never were. They are are heavily cited. --C S (Talk) 16:46, 20 April 2006 (UTC)

Looks much better to me now. Dmharvey 16:49, 20 April 2006 (UTC)

My problem with adding this is that it makes arXiv sound even more reliable. The article currently doesn't say anything about the papers not being peer reviewed, nor does it say anything about the inclusion of papers by various crackpots (the late Caroline Thompson, for example). I'm not sure how much of a problem this is outside Wikipedia, but people inside Wikipedia often try to mislead unknowing editors by giving references to such papers on arXiv. I would add this, but I am unsure of how to do so in an unbiased and elegant manner. --Philosophus ^T 19:50, 7 May 2006 (UTC)

Good point - this wasn't mentioned in the article. I've made an attempt, hopefully reasonably unbiased. Deuar 19:58, 9 May 2006 (UTC)

I believe that Deuar's attempt to clarify the raised concerns has caused bigger problems. The edit basically consists mostly of opinion, albeit informed opinion. I don't believe it adheres to either WP:NPOV or WP:Verifiability. If the concern is that the article does not mention explicitly it is not peer-reviewed (although it is implicit in the second sentence), we can just say that. We don't need to explain why this unreviewed eprint archive may not be as reliable as a peer-reviewed medium, or how certain articles may be reliable given certain conditions. Let's just stick with the facts and explain the whole idea of journal overlays (a major way refereed papers get on arXiv), endorsements, and so forth. Those are important pertinent issues and can easily be explained in a neutral manner. --C S (Talk) 16:23, 14 May 2006 (UTC)

Well, true, I just wrote thoughts that occured to me from experience. I'm not sure which parts you consider non-neutral but feel free to improve it! Deuar 18:26, 14 May 2006 (UTC)

While the text that you removed was somewhat opinionated, the new version doesn't mention the problems at all. It should be easy to cite a crank paper or two in arXiv, and mention that there are such papers there. --Philosophus ^T 01:02, 15 May 2006 (UTC)

I agree that it should be mentioned as it is one of the things which pops up very frequently in discussions about the arXiv. I tried to find a neutral formulation, supported by a reference. -- Jitse Niesen (talk) 02:28, 15 May 2006 (UTC)

Jitse, your edit falsely implies that the endorsement system has to do with ensuring correctness. As my edit should have made clear, the endorsement is NOT to ensure correctness but only that the submissions are appropriate and relevant to the research going on in the specified areas; also, for some reason, the comment in the first paragraph about the papers being eventually published is repeated (with the unverifiable "most"). I note that the Jackson article explicitly mentions the lack of statistics about which percentage of eprints gets published (while mentioning one informal test of 100 papers in hep-th by Kuperberg). It does mention that a majority of papers are submitted for publication though.

I will change it back to be correct, while keeping your citation and modifying your comment about the concerns, as the article doesn't quite make the point I believe you want, Jitse. The Jackson article mentions that people that use the arXiv are not concerned by the lack of peer-review, and that "junk", crank-type papers are actually infrequent. So while we could mention as Philosophus suggests, that such papers do exist on the arXiv, we would need to mention their rarity.

Anyway, I've edited based on these comments. --C S (Talk) 05:34, 15 May 2006 (UTC)

1. "Jitse, your edit falsely implies that the endorsement system has to do with ensuring correctness." — Yes, my formulation was sloppy, sorry.
2. "the comment in the first paragraph about the papers being eventually published is repeated" — My plan was to remove it from the first paragraph, but I forgot that. I think it is logical to put the facts "some papers are not published in peer-reviewed journal" and "the majority are" next to each other. I edited to this effect.
3. "the article doesn't quite make the point I believe you want" — I think it is just that I was not very clear in expressing what I had in mind. I agree with your edits, modulo details.

I do have the impression that the endorsement system was put in place as an (incomplete) replacement for peer-review (but, working in one of the areas that does not use the arXiv much, I may well be wrong). For that reason, I put the "problems with no peer review" paragraph before the paragraph describing the endorsement system. -- Jitse Niesen (talk) 06:11, 15 May 2006 (UTC)

I'm not sure what you mean by incomplete replacement, but it's never appeared to me that the endorsing was designed to be anything more than a very minimal barrier to keep out incredibly ridiculous papers, in order to minimize "cruft" buildup and reduce workload for those working for the arXiv. The requirements to be an endorser are fairly lax, and if you really wanted to, I'm sure you could find someone to endorse your latest proof of Fermat's Last Theorem :-) My impression is that endorsement is designed to keep out some of the cranks, but it certainly was never meant to seriously reduce the number of errors in papers on the arXiv. Anyway, the new order is fine, as it does suggest the right things. --C S (Talk) 07:06, 15 May 2006 (UTC)

The recent edits look like a fine piece of work to me. Deuar 14:25, 15 May 2006 (UTC)

The constraints imposed on endorsers are not lax. The arXiv help page about endorsement ends with this warning: "We reserve the right to suspend a person's ability to endorse for any reason. If you endorse a person who makes an inappropriate submission, we may suspend your ability to make endorsements. If you feel uncomfortable about endorsing an author for any reason, don't do it -- ask the person to find another endorser." For this reason it is very difficult for people outside of academia to find endorsers even for quite reasonable papers. Weburbia (talk) 08:14, 27 June 2010 (UTC)

I was asked to endorse on arXiv 4 times and I got to say yes only once. Twice the author did not respond to my request for a copy of the article and once it was a clear crank. Jmath666 (talk) 05:46, 17 September 2010 (UTC)

Archive Freedom

The website http://archivefreedom.org/ is critical of arXiv claiming that it operates a blacklist. At least one of the people mentioned Dr. Peter Rowlands is a university lecture in a physic department which has been given the top rating for research in the UK (5A). The transcript [1] indicates this is not just obvious cranks being excluded.

Normally I would not think of including such a site in wikipedia, but it does seem on topic for this article. --Salix alba (talk) 21:52, 25 May 2006 (UTC)

I don't think anyone has disputed this should belong, which is why it has remained :-) So I'm a little puzzled as to what brought on this comment. --C S (Talk) 01:00, 28 May 2006 (UTC)

My mistake, it was mention on Talk:Pseudoscience, I quickly scanned this article but did not notice it, at the end of a paragraph. --Salix alba (talk) 10:43, 28 May 2006 (UTC)

Hi, can someone cite the exact paragraph in Talk:Pseudoscience that deals with the exclusion of honorable scientists ?

By the way: there lies an (hopefully;) ) unintended sarcasm in the fact that if someone seeks information about archive freedom they get all kinds of results, but archivefreedom.org is (at least) not on the first three pages (although it starts with an a). What is even worse, is if someone writes archivefreedom as one word, the sole result is Pseudoscience ! This suggests archivefreedom = Pseudoscience, which I believe no one here intended . Phi0618 (talk) 11:26, 21 November 2007 (UTC)

Thank you, someone took care of the archivefreedom result, it now shows no results.

I'm going to make my first edit, just a mention of archivefreedom in the title of the section that deals with it. Phi0618 (talk) 13:24, 21 November 2007 (UTC)

The balancing mention of possible blacklisting was deleted by a non registered user. I would therefore like to encourage a discussion and vote by other Wiki's on the subject. Phi0618 (talk) 16:39, 22 November 2007 (UTC)

I just noticed C S 's comment that no one disputed it should belong, so I'll boldly put it back in, justified by the fact that every statetment can be found in the reference. —Preceding unsigned comment added by Phi0618 (talk • contribs) 16:45, 22 November 2007 (UTC) Phi0618 (talk) 19:11, 22 November 2007 (UTC)

Also,, there is another kind of 'censorship' you cannot publish papers on Arxiv if you do not have an academic affiliation to an University no matter if your paper is correct or not --85.85.100.144 (talk) 13:17, 20 January 2008 (UTC)

I think that's no longer the case. Brian Josephson (talk) 15:24, 17 March 2023 (UTC)

Perhaps someone could add a page to describe viXra, the archive established by and for authors who find they cannot submit to arXiv becuase they do not have an endorser or have problems with the arXiv moderation policy. I can't do it myself due to conflict of interest but I would use the Talk section if necessary to ensure accuracy or make suggestions. Weburbia (talk) 08:23, 27 June 2010 (UTC)

External links

Most of the external links should be made into references cited in an appropriate place in the text. Jmath666 17:15, 8 September 2007 (UTC)

Does Arxiv provide any economic benefits?

Given the time and effort that authors will go to in order to publish or create the results which they then go onto publish in certain peer-reviewed journals, together with those papers which they place on Arxiv, are there some economic mechanics behind ArXiv?

Are there plans for some type of incentive system which would enable ArXiv members to benefit or profit from the work that they publish? It would seem common sensical that work which might have practical application for economic betterment (whilst also being of a mathematical nature) would be well suited to presentation on an ArXiv equivalent site (assuming, of course, that we would like to keep ArXiv itself open source and free for all). Prizes could then be given to economically beneficial papers - or papers could be recommended in some way.

It would seem to me that much of the ArXiv information could find application to various uses – and hence that it should be possible to relate at least some ArXiv articles to some type of economic value (sorry if this brings ideas of impact factors to mind).

CountNihilismus 23:01, 8 November 2007 (UTC)

Of course. ArXiv authors gain because their work gains wider exposure than in a peer-reviewed journal alone: clearly before the publication, but also after the publication the work is accessible conveniently to those whose institutions do not have subscriptions to the journal, although only as a preprint, which may or may not be the same as published in the journal. (ArXiv is way better than putting a preprint on one's own or institution's website.) Wider and stable exposure may lead to the work being more widely used and cited, which is a factor in the author's academic advancement and standing and affects promotions, raises, and grants. This an economic benefit. Unfortunately ArXiv's attitude to search engines and robots makes this benefit less than what it could be.

ArXiv gains because higher usage makes it more likely to get more funding from sponsors such as NSF. Research community gains, because they get free access to the information. Publishers lose because ArXiv decreases the pressure on the institutions to buy subscriptions. That's why the publishers often do not like ArXiv and try to lock it out in their copyright transfer agreements.

Jmath666 (talk) 20:02, 28 November 2008 (UTC)

Copyright

I am trying to find pictures of the missing Fields Medallists that are acceptable by Wikipedia's standards. A few of them (like Vladimir Voevodsky's, for example) can be found in pdf files in the arXiv. Does anybody know whether arXiv is free content or at least "fair use" according to Wikipedia? —Preceding unsigned comment added by 164.41.88.203 (talk) 16:31, 28 November 2008 (UTC)

See http://arxiv.org/help/license — arXiv allows authors to make their papers public domain, CC, or GFDL, but the default is merely to grant a license to the arxiv to publish them. That's not good enough to allow us to re-use the content here. —David Eppstein (talk) 17:34, 28 November 2008 (UTC)

Thank you very much for the quick response. Are you quite sure that arXiv's conditions for storing the e-prints are not good enough for Wikipedia? From what I have seen there, they do not say the authors can make their e-prints CC, rather that they must if their articles are to be stored there. And I understand that CC is good enough for Wikipedia (see, for example, the picture of Alexander Grothendieck, which was not uploaded by me). The only alternative - which seems to be the preferred one not to conflict with later publication standards - to a CC is the arXiv-specific license at http://arxiv.org/licenses/nonexclusive-distrib/1.0/license.html, which, I agree, is not enough for Wikipedia. So, there is still a chance that, for example, the picture of Voevodsky's will be allowed at Wikipedia, if we can work out which license the submitter has chosen. Anyway, thanks for finding that page for me. It seems that the only Fields Medallist with a picture there is Voevodsky, so it will be easy to check.

Incidentally, the non-exclusive redistribution license was created in 2004, so I think it is a pretty safe assumption to say that e-prints prior to that date (like the one with Voevodsky's picture) had to be submitted with a CC license. Should I risk uploading the image? Or ought I to contact the submitter or arXiv itself before any such attempt? —Preceding unsigned comment added by 164.41.88.203 (talk) 18:30, 28 November 2008 (UTC)

If it doesn't say that it's freely licensed (e.g. in the comments section of the abstract page), I wouldn't trust that it's freely licensed. The current situation is that it may use the arxiv license (not free), OR creative commons, OR be public domain: it's a disjunction, not a conjunction. My recollection is that prior to 2004 all papers were released under conditions similar to the current arxiv license. —David Eppstein (talk) 18:43, 28 November 2008 (UTC)

That's also my recollection. Besides, the arXiv is way older than the Creative Commons so it can't have been using a CC license from the start. -- Jitse Niesen (talk) 19:10, 28 November 2008 (UTC)

Which license should I use in arxiv in order to guarantee that the plots can then be re-used in wikipedia? Alessandro.de.angelis (talk) 11:44, 28 October 2016 (UTC)

CC-BY or CC-BY-SA. CC0 would also work but I wouldn't recommend it for a paper. Don't use CC-BY-NC-SA or the default nonexclusive arXiv license. —David Eppstein (talk) 18:10, 28 October 2016 (UTC)

Underlying Software / Hardware

Information on what hardware and software underlies ArXiv seems to be sparse on the Internet and in this article. The ArXiv help files occasionally reference something called "AutoTeX" (e.g. at http://arxiv.org/help/submit_tex) but searching the web for this term yields unrelated results such as http://www.surf.nuqe.nagoya-u.ac.jp/~nakahara/Software/AutoTeX/index-e.html

Can anyone locate a description of the platform on which ArXiv operates, whether it is made available, and if so, under what sort of license? — PaulKishimoto (talk) 02:58, 20 December 2008 (UTC)

I can ask on mod-admin if you think that would be helpful. But to add anything to the article, we're going to need to find a reliable source. —David Eppstein (talk) 03:15, 20 December 2008 (UTC)

Controversy section whittled down to nothing

Could we at least have a paragraph or so on Vixra ? I'd do it myself, but I've had it with trolls. Robma (talk) 14:09, 22 November 2009 (UTC)

Does Vixra pass WP:WEB? I don't see anything about it in Google news archive. How could we source such a paragraph? —David Eppstein (talk) 17:22, 22 November 2009 (UTC)

I noticed there was a healthy controversy section recently but it has been incrementally whittled down to nothing. Are there trolls here trying to paint an unbalanced view? I welcome editors to reconstitute it. Otherwise I'm going to research the past content and reconstruct it. Please someone explain why it was wiped out. 68.129.183.213 (talk) 23:15, 25 April 2018 (UTC)

It was whittled down because the sources it was based on were not reliable. In the most recent removal, the editor who removed the material wrote "Removed unreliable source. The guy is a PhD student and his website reveals him to be a grade-A crank". Do not re-add without proper reliably-published independent sourcing. —David Eppstein (talk) 02:27, 26 April 2018 (UTC)

SnarXiv

A deletion discussion for a parody site called "SnarXiv" resulted in a decision that material about that site should be merged here, to arXiv, with no notification to editors of this article nor any attempt to create a consensus here that such a merge is appropriate. I believe it is inappropriate (the material is not notable and has too little significance with respect to the arXiv itself to be mentioned, per WP:UNDUE) and have reverted the change. For an ongoing discussion of this issue, see Wikipedia:Deletion review/Log/2014 March 8. —David Eppstein (talk) 16:45, 8 March 2014 (UTC)

Unfortunately, SnarXiv still redirects here, confusing links like the one in SCIgen#See also. 129.93.4.34 (talk) 17:06, 10 July 2014 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on ArXiv. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20050505044157/http://www.public.iastate.edu/~gerrymck/arXiv.org.pdf to http://www.public.iastate.edu/~gerrymck/arXiv.org.pdf

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 07:15, 8 July 2017 (UTC)

First Line of the Article

This talk entry is to address the first line of the article. In particular, to address the phrase "peer review". According to peer review: "Peer review is the evaluation of work by one or more people of similar competence to the producers of the work (peers)." This is what the committees that review arXiv submissions do. Their reviews determine such things as if the work is deemed "scientific" in the opinion of the corresponding arXiv committee. They also provide written feedback to submitters on individual aspects of the submissions such as citations. For example, a submitter was told to remove a citation to a viXra publication as a condition of having his arXiv submission accepted. These examples are clearly "evaluation of work by one or more people of similar competence". 173.56.74.61 (talk) 00:30, 11 December 2017 (UTC)

Moderators merely check whether submissions meet some very basic standards (e.g. do they have references) and whether they are on-topic for their category. In most cases this entails a quick scan of the abstract of the article without even reading the article itself. There is very little quality or correctness checking, except maybe in some cases rejecting obvious crankery, plagiarism, or copyright violation. It is not peer review in the usual sense of that phrase. (I am speaking from personal experience here, but not in any official capacity for the arXiv; I am a moderator there.) —David Eppstein (talk) 01:45, 11 December 2017 (UTC)

I agree with the article that arXiv is not peer-reviewed. Most papers are accepted through a two level process of endorsement plus an automated check. According to how the process has been described, once authors have been endorsed for some topic categories, their subsequent submissions are not checked by any human peer or moderator unless they are flagged up by the automated filters. Do you have a reliable source to confirm that removal of a vixra link has been used as a condition for acceptance? I don't doubt that a reference to vixra would trigger the filters, but there are papers on arXiv that do have references to papers on viXra. Weburbia (talk) 07:19, 11 December 2017 (UTC)

David, you provide a claim on what "moderators" do yet you casually side-stepped the fact that an arXiv submitter was instructed to remove a reference to a viXra article as a condition of the submission being accepted for publication by arXiv. David, are you saying that the testimony of Philip Gibbs is false? 173.56.74.61 (talk) 12:56, 14 December 2017 (UTC)

I can confirm that what I wrote in that essay is true, but whether that qualifies for inclusion under relevant wikipedia policies is something I leave to others to judge. I understand that it's not simply a case of whether someone thinks it is true or false.Weburbia (talk) 17:04, 14 December 2017 (UTC)

Additional evidence of peer review by arXiv is the article "Does the arXiv censor submissions?" published by Sabine Hossenfelder on January 28, 2016. In this article she describes a paper covering backreaction relating to black holes that was rejected by arXiv. She stated "Ok, so the paper is wrong. But should it have been rejected by the arXiv? I don't think so. The arxiv moderation can't and shouldn't replace peer review, it should just be a basic quality check, and the paper looks like a reasonable research project." David Eppstein, please comment on this peer review event as well as the one I pointed out above on 14 December 2017 (UTC) that is described in the article "Open Peer Review to Save the World". 173.56.74.61 (talk) 20:45, 16 December 2017 (UTC)

Look, if you don't understand that summarily filtering out blatant crankery and plagiarism is a different thing than actual peer review, or don't want to understand, I can't help you. And when you say "article, published" what you actually mean is "blog, posted"; that's not something we can use as a source for any content here. Stop pinging me for this nonsense. —David Eppstein (talk) 07:20, 17 December 2017 (UTC)

David Eppstein, I'm not sure why you are reacting this way. I am making an honest attempt to work out our disagreement on the fact that arXiv peer reviews submissions. Your response does not address either of the two instances of peer review by arXiv that I pointed out. 173.56.74.61 (talk) 23:48, 18 December 2017 (UTC)

I'll have a go at it, then. Peer review is done by researchers who are specialists in the topic of the paper, or at least a small subfield to which the paper belongs. They carefully read the entire paper and usually have a good understanding of the material, the methods, and the relevance of the research. Using their expertise, they comment on mistakes, flawed methods, unclear/irrelevant content, omissions, failures to discuss existing literature, etc, and they may suggest ways to improve the paper. It can take a full workday to review a single paper properly, and a referee report can be several pages of text. The arXiv moderators have to handle circa 500 submissions each workday. Just based on the size of the team (about 170), they cannot have a specialist on each topic across the wide range of arXiv areas, and are therefore not 'peers' in the sense of peer review. Furthermore, they have to moderate about three submissions per person per workday on average. It's quite clear that they will only spend at most 5-10 minutes per paper, and maybe even less (see David Eppstein's description above). This is vastly more superficial than peer review. Your examples are fully consistent with the superficial moderation, and are no evidence at all that the arXiv moderators are somehow spending all their time performing peer review. Gap9551 (talk) 00:46, 19 December 2017 (UTC)

Just a minor correction: it is more than 3/day. Today's email for me (so far) includes 14 arxiv-moderator emails, most of which involve asking me to look at a paper and make a decision on it. Even 10 minutes per paper would be too much at that volume. —David Eppstein (talk) 01:04, 19 December 2017 (UTC)

Thanks. Gap9551 (talk) 02:10, 19 December 2017 (UTC)

Gap9551, your input is much appreciated. Please explain how/why you believe that this has anything to do with my arguments thus far: "and are no evidence at all that the arXiv moderators are somehow spending all their time performing peer review" 173.56.74.61 (talk) 01:51, 19 December 2017 (UTC)

Huh? Are you saying your arguments so far are not about arXiv doing peer review? If so, please tell me what we are discussing here. Gap9551 (talk) 02:10, 19 December 2017 (UTC)

By the way, arXiv has posted over 1.3 million papers so far. Even if there have been a few cases of a moderator going beyond their normal duty and actually doing peer review on a submission, that does not mean that this is their standard practise. Gap9551 (talk) 02:13, 19 December 2017 (UTC)

Gap9551, it appears that you have not yet read this section. Please read it, the issue at hand was stated clearly by myself above on 14 December 2017 (UTC) and 16 December 2017. In the former an arXiv peer reviewer instructed a submitter, in the written peer review feedback to the submitter, that the submitter must remove a reference to a viXra paper in order for the arXiv submission to be accepted. In the latter Sabine Hossenfelder provides testimony regarding a submission that was peer-reviewed. To be succinct: my claim is that arXiv peer-reviews submissions. Now that I've answered your question, please answer mine: Please explain how/why you believe that this has anything to do with my arguments thus far: "and are no evidence at all that the arXiv moderators are somehow spending all their time performing peer review" 173.56.74.61 (talk) 02:20, 19 December 2017 (UTC)

You are saying, my claim is that arXiv peer-reviews submissions and Please explain how/why you believe that this has anything to do with my arguments thus far: "and are no evidence at all that the arXiv moderators are somehow spending all their time performing peer review". I'm assuming you have noticed that both phrases contain the key words "arXiv" and "peer review", so it does not make sense to suggest that my statement has nothing to do with your arguments. At the same time you ignored my Even if there have been a few cases of a moderator going beyond their normal duty and actually doing peer review on a submission, that does not mean that this is their standard practise, which is highly relevant to the issue you bring up in your last comment, regarding removing a reference (who knows if that even happened, but that's not the point, I'm willing to assume it did). Your unwillingness to be open to discussion forces me to leave this "conversation", because I am wasting my time. Do not introduce your so far unsubstantiated claim in the article without consensus. Gap9551 (talk) 02:46, 19 December 2017 (UTC)

Site icon (favicon)

I suspect that the site icon is a reference to a design used by The Wildhearts in 1992. Image:WildiesDontBeHappy.jpg but don't have a reference. At that time there was so little graphic art available on the web people used to reuse icons from the small number openly available]] Elseware (talk) 08:21, 3 April 2019 (UTC)

ArXiv has used a skull-and-crossbones theme since its founding (as xxx.lanl.gov) in 1991. —David Eppstein (talk) 16:56, 3 April 2019 (UTC)

Quora (not quite an RS) has some useful informal information on the icon history.[2] I was originally a typical skull and crossbones, reflecting both the xxx of the web site and (my own personal recollection, definitely unreliable) the pirate nature of passing around preprints without publisher approval. The smiley face came after, making it less offensive to physicists w/o a sense of humor. --{{u|Mark viking}} {Talk} 17:11, 3 April 2019 (UTC)

"Obsolete" source

I just removed an "obsolete source" tag without realizing that I was at least the third editor to do so. To me, the matter is clear: it's not "obsolete," it's history. Having a more recent source on the same topic would be nice, but knowing what the past situation was like continues to be important. As David Eppstein's edit summary points out, We do *not* need a better source for an opinion that already has a date and attribution. That source is fine for the claim that, *in 2002, the AMS stated* that these instances were surprisingly rare. We would need a better source to say that they're still rare, but we're not saying that. XOR'easter (talk) 15:38, 8 October 2019 (UTC)

Fully agree with XOR'easter and David Eppstein here. Headbomb {t · c · p · b} 21:33, 8 October 2019 (UTC)

Academic papers or scientific papers?

I changed "scientific papers" of ArXiv for "academic papers" to avoid confussion, because whether mathematics is or not a science is strongly debated. On the other hand, all ArXiv preprint papers are academic, so with the word academic all problems are solved.

Extended content

There exists a strong debate on whether mathematics is or not a science. It depends on the conception of science and mathematics each one has. Of course, what constitutes science is also hihgly debated (the demarcation problem). Anyhow, science is not a natural object that exists independently of humans. Science is a human creation, and so what constitutes science is decided by humans. So the question of whether mathematics is or not science is a bit subjective, in the sense that what constitutes science is decided by us humans. Among the main reasons to NOT consider maths a science are its non-empirical method (in contrast with science) and its object of study. While scientists test their hypotheses by observing the natural world, mathematicians work on abstract entities. There is also a debate on whether mathematical objects and numbers are part of nature or an abstraction created by humans. Many university departments have faculties of "Mathematics AND science" (implicitly suggesting a separation of both concepts): http://www.deltastate.edu/artsandsciences/mathematics-and-sciences/ https://brocku.ca/mathematics-science/ https://smt.sutd.edu.sg/faculty/ https://ems.tcd.ie/ https://www.lyon.edu/math-science-faculty https://www.aubg.edu/mathematics-and-science-department https://www.cedarville.edu/Academic-Schools-and-Departments/Science-and-Mathematics.aspx https://www.muw.edu/case/scimath https://www.everettcc.edu/programs/math-science Some English dictionaries do not define Mathematics as a science. Among those dictionaries that do not define mathematics as science are: Dictionary.com McMillan Dictionary Many authors do not consider mathematics to be a science: "Mathematics is not science, and its turths must not be understood as empirically verifiable". -Alan Bishop. Page 54 of the book Mathematical Enculturation: A Cultural Perspective on Mathematics Education. https://books.google.es/books?id=9AgrBgAAQBAJ&pg=PA54&dq=mathematics+are+not+science&hl=es&sa=X&ved=0ahUKEwjRlYbvm9zmAhWKGBQKHXpACqwQ6AEIeTAI#v=onepage&q=mathematics%20are%20not%20science&f=false "mathematical knowledge isn’t the same as empirical knowledge, and consequently mathematics is, indeed, a different way of knowing from science." - Massimo Pigliucci. Are there "other" ways of knowing? (2014) Philosophy now. https://philosophynow.org/issues/102/Are_There_Other_Ways_of_Knowing "Nonscience traditionally includes not only pseudoscience and metaphysics but also logic, pure mathematics, and other subjects that cannot be tested against experience, including the normative topics studied in value theory." -Thomas Nickles. The Problem of Demarcation. Page 104 of Chapter in the book Philosophy of Pseudoscience (2013). Many articles of science education also explicitly separate science from mathematics: "The more successful professional development programs were not simply courses in mathematics or science" -MM Kennedy. Form and Substance in Mathematics and Science Professional Development. In Eric: Institute of Education Sciences. (1999) https://eric.ed.gov/?id=ED435552 "they realized learning science and mathematics does not have to be stressful" -DA Huinker. Preparing efficacious elementary teachers in science and mathematics: The influence of methods courses. In Journal of Science Teacher Education (1997) https://www.tandfonline.com/doi/abs/10.1023/A%3A1009466323843?journalCode=uste20 Connecting Science and Mathematics Instruction: Pedagogical Context Knowledge for Teachers. -Frykholm. In School Science and Mathematics. (2010) https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1949-8594.2005.tb18047.x The webpage "Understanding science" is founded by the National Science Foundation and attempts to both define and describe science. Even in Understanding Science it is said that "math is closely related to science" (suggesting that both domains can be considered similar but different): https://undsci.berkeley.edu/article/mathematics

In sum, I am not trying to defend one reasoning (mathematics is science) or the other (mathematics is not science). I am just trying to illustrate that it is a highly debated issue. Therefore, I believe it is better for the article to simply talk about "academic papers" of ArXiv. ALL articles on ArXiv are academic, therefore there is nothing misleading about it. Whether all of them are scientific or not is a mather of dispute (mathematics), hence I think it is useful to change "sceintific articles" for "academic articles" to avoid the problematics of assuming that mathematics is universally or generally considered science (when it is debated). James343e (talk) 2:10, 30 December 2019 (UTC)

"Scientific" is fine for a short description. It is true, as far as it goes, whereas "academic" is too broad. XOR'easter (talk) 03:03, 30 December 2019 (UTC)

Arxiv describes itself as a "free distribution service and an open archive for scholarly articles."[3]. How about using "scholarly"? The term encompasses math, science, engineering, finance, economics, etc., as topics, and academic, corporate, and private sources. --{{u|Mark viking}} {Talk} 04:18, 30 December 2019 (UTC)

"Scholarly" is just as broad as "academic", and thus too broad. And no: Arxiv does not describes itself just as a "free distribution service and an open archive for scholarly articles", but much more precisely as a "free distribution service and an open archive for scholarly articles in the fields of physics, mathematics, computer science, quantitative biology, quantitative finance, statistics, electrical engineering and systems science, and economics". This is much too long, and "scientific" is fine for a short description. Sapphorain (talk) 11:09, 30 December 2019 (UTC)

I must disagree. Scholarly and academic are two different concepts; scholarly is an approach to knowledge, whereas academic is a kind of institution. Engineering is not usually considered science, nor is quantitative finance. --{{u|Mark viking}} {Talk} 12:05, 30 December 2019 (UTC)

What about "technical"? XOR'easter (talk) 16:42, 30 December 2019 (UTC)

"Technical" is also a good descriptor that I would support. --{{u|Mark viking}} {Talk} 20:03, 30 December 2019 (UTC)

Much fuss about nothing. I'm fine with either scientific or academic. Technical or scholarly is pedantic. Headbomb {t · c · p · b} 20:06, 30 December 2019 (UTC)

Defunct links

Paul Ginsparg's "The global-village pioneers" is no longer available at the sourced URL (directs to an error 404 page). Should the link to this article be updated, or is it relevant to cite here in the first place? Paxrei (talk) 21:02, 10 October 2020 (UTC)

Updated. XOR'easter (talk) 21:51, 10 October 2020 (UTC)

"ArKiVe" listed at Redirects for discussion

An editor has identified a potential problem with the redirect ArKiVe and has thus listed it for discussion. This discussion will occur at Wikipedia:Redirects for discussion/Log/2022 July 10#ArKiVe until a consensus is reached, and readers of this page are welcome to contribute to the discussion. Headbomb {t · c · p · b} 14:47, 10 July 2022 (UTC)

‏source isn't reliable

I referring here https://www.space.com/dark-energy-2nd-big-bang-research

on this astronomical website source of this study was mentioned: https://arxiv.org/abs/2302.03710

We all know that arXiv is not peer reviewed…I do not think it is wise to add a preliminary study to the encyclopedia...it is still not reliable Clerk18890 (talk) 20:38, 25 March 2024 (UTC)

This talk page is for discussions of how to improve the Wikipedia article about arXiv, not for the use of arXiv elsewhere on Wikipedia. That said, arXiv:2302/03710 is the same as doi:10.1103/PhysRevD.107.083533, a reliably published paper in a major journal. —David Eppstein (talk) 20:51, 25 March 2024 (UTC)

Sorry, I misunderstood this page Clerk18890 (talk) 06:45, 25 May 2024 (UTC)

An external link

There has been an objection by ElKevbo (talk · contribs) on putting a link to a post about a submission to arxiv. This incident may not be significant enough to be properly discussed in the article but I don’t see why it is necessary to suppress having a link to the post about the incident. In Wikipedia, we surely don’t want to appear censoring some submission incidents. Taku (talk) 12:28, 26 July 2024 (UTC)

We try to be very selective about what we include in the "External links" section of every article. I don't think that the link you added to a blog post meets our criteria for inclusion. If the "incident may not be significant enough to be properly discussed in the article" then why in the world would we include a link to it in the "External links" section? ElKevbo (talk) 22:13, 26 July 2024 (UTC)

Indeed, a random blog post isn't EL material. Headbomb {t · c · p · b} 22:24, 26 July 2024 (UTC)

It seems what is relevant here is. "11. NOBLOGS Blogs, personal web pages, and most fansites (negative ones included), except those written by a recognized authority", in particular, the last part of this sentence. The n-category cafe is a fairly prominent website in the category theory community. I am not convinced either way that the incident is significant or is insignificant but, for me, the prominence of the website was enough. But I guess the website may not be well-known outside the category-theory community. Anyway, since the consensus seems against the inclusion, I will remove the link then. —- Taku (talk) 05:46, 27 July 2024 (UTC)

Language link to Thai article

th:arXiv is now available for Thai language. Would be nice if we can interlink this with other languages. Thank you.

(The Wikidata article for arXiv is locked. So I can't update it and it can't be auto-updated from the Translation tool). -- bact' (talk) 10:57, 4 November 2024 (UTC)

I don't see any evidence that the data item is locked, but it does require autoconfirmed on wikidata. Anyway, I added the link. —David Eppstein (talk) 17:47, 4 November 2024 (UTC)

Thank you -- bact' (talk) 05:04, 5 November 2024 (UTC)