Not signed in (Sign In)

Not signed in

Want to take part in these discussions? Sign in if you have an account, or apply for one below

  • Sign in using OpenID

Discussion Tag Cloud

Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.

Welcome to Math2.0
If you want to take part in these discussions either sign in now (if you have an account), apply for one now (if you don't).
    • CommentRowNumber1.
    • CommentAuthornad
    • CommentTimeSep 28th 2012

    there had been some hoax papers, the most prominant probably being the famous Sokal hoax. I was wondering to what extend a hoax scheme had been discussed in the academic community.

    what do I mean by a hoax scheme (note this is just a funny gedanken experiment): lets imagine an institution like the AIP or the AMS secretly assigns hoax authors. That is distinguished members of such societies secretly publish hoax papers under the auspices of these societies. Citing such hoax papers would of course be some kind of sign that there had been something overlooked, i.e. the “citation index” of official hoax papers could be interpreted as a quality sign and thus serve as a possibilty to sort “bad” from “good” papers.

    obvious disadvantages of such an approach are of course the lost effort one had to make to go through such papers and identify it as a hoax, or in other words information retrieval could get even more difficult with hoax papers from distinguished authors then that it gets more difficult by going through non-hoax, but nonetheless useless articles. moreover a crucial question is of course also how long a hoax can be kept as a hoax.

    • CommentRowNumber2.
    • CommentAuthordarij grinberg
    • CommentTimeSep 28th 2012

    What exactly is the goal? Identifying bad papers by the presence of hoaxes in the list of references? I fear that this minefield will blow up randomly rather than picking the right targets. One of the unwritten rules of mathematical writing is that it’s perfectly normal to have read about 10% of the stuff you are citing, and “better too much than too little” is the way to go in lists of references. It is much more embarassing to miss a relevant reference than to have some irrelevant ones cluttering up the list, so people regularly over-cite. I think the references aren’t a quality criterion of any kind for a paper.

    Sokal worked differently: it was about getting the hoax published, not getting it cited. Even with that, I am unsure that this would work on a regular basis – at some point, editors would start checking all incoming papers for signs of irony and trolling rather than for actual quality, as the latter is harder to achieve. At the end of the day, all that will happen is that a few journals will have been publicly shamed and publishing a paper will have become a bit more expensive in time and money. Not sure if there will be any lasting impact…

    • CommentRowNumber3.
    • CommentAuthornad
    • CommentTimeSep 28th 2012
    • (edited Sep 28th 2012)

    it’s perfectly normal to have read about 10% of the stuff you are citing

    there were times when this ratio was higher. I understand that not everybody has access to all articles, and that it would look odd not to include the originator of a subject, but still a good citation list can be a quality criterion. even for citations which are rather included for completeness reasons than that they are needed for understanding an article -since those citations are intentended to give an overview over a subject and hoaxes are not really helpful.

    So it could be still interesting to see how much hoaxes are distributed. And this doesnt need to be on a regular basis, but could be e.g. one test.

    I think actually that it would be helpful to identify in a citation how much you have read/understood of an article, but thats another thing.

    and as said this is just a gedankenexperiment.

    • CommentRowNumber4.
    • CommentAuthordarij grinberg
    • CommentTimeSep 28th 2012
    • (edited Sep 28th 2012)

    I think actually that it would be helpful to identify in a citation how much you have read/understood of an article, but thats another thing.

    Definitely. But this seems even harder than making everybody stop publishing in Elsevier. Saying publicly that one has not read a paper will always have a connotation of either “I don’t care for the author’s work” or “the paper is unreadable”.

    • CommentRowNumber5.
    • CommentAuthornad
    • CommentTimeSep 28th 2012
    • (edited Sep 28th 2012)

    Saying publicly that one has not read a paper will always have a connotation of either “I don’t care for the author’s work” or “the paper is unreadable”.

    actually I found it is increasingly the case that I tried to read a paper and unfortunately found that it is definitely unreadable. this by the way may hold also for some talks.

    • CommentRowNumber6.
    • CommentAuthorScott Morrison
    • CommentTimeOct 1st 2012

    This seems to be a solution in search of a problem.

    • CommentRowNumber7.
    • CommentAuthornad
    • CommentTimeOct 1st 2012

    This seems to be a solution in search of a problem.

    who is searching a problem?

    do you mean the discussion about a refinement of citation indexes with respect to citing authors evaluations could be a thesis subject, like for a BA in the social sciences?

    With definitely unreadable I mean articles, which seem to me to be without any for me reasonable content. There are different sorts of unreadable articles. Like there are articles, where I think that they may be worthwhile to read after I I learned more about a subject, or articles, where I need to get more accustomed to the authors style and way of thinking. But there are articles and authors where I find even after several attempts that they still don’t make sense to me and at one point I may even stop reading the articles of this author. For a lot of articles one actually can still start to guess what the authors could mean and then if this looks somewhat consistent, assume that this what one came up with is what the authors could have meant…but thats tricky of course and it takes a lot of time.

    But there are actually also authors who I understand almost immediately.

    • CommentRowNumber8.
    • CommentAuthornad
    • CommentTimeOct 1st 2012
    • (edited Oct 1st 2012)

    I want to add that this problem of guessing of what the authors could have meant extends also to information technology. And it can be very helpful if you can ask someone. Like if you have descriptions like “in the upper right panel on the lower left corner you can find a pop down menu with a scrollable window box that holds a set of quick links to a menu which contains the overview front screen panel” and these descriptions aren’t fully up to date, because the the scrollable window box has meanwhile changed into the upper right pop down menu then my husband actually usually knows where to look. My theory is that this is because he may think somewhat similar to the way the programmers think.

    • CommentRowNumber9.
    • CommentAuthornad
    • CommentTimeOct 10th 2012

    I asked:

    I was wondering to what extend a hoax scheme had been discussed in the academic community.

    I ask this because I think betting out prices on finding hoax mines could be a new income game, eventually leading to new math professions like hoax-container-deposit-BOOBAKI-garbage-collectors.

    • CommentRowNumber10.
    • CommentAuthornad
    • CommentTimeOct 19th 2012

    One would however probably need to guarantee some anonymity for the players and their findings in order to avoid possible troubles, like after identifying a bullshit paper wrongly as a hoax.

    • CommentRowNumber11.
    • CommentAuthorzskoda
    • CommentTimeOct 26th 2012
    • (edited Oct 26th 2012)

    Darij, 2 said

    One of the unwritten rules of mathematical writing is that it’s perfectly normal to have read about 10% of the stuff you are citing, and “better too much than too little” is the way to go in lists of references. It is much more embarassing to miss a relevant reference than to have some irrelevant ones cluttering up the list, so people regularly over-cite.

    I would not say this is (an unwritten) rule at all. In many tradition it is more of the opposite. I think that Vladimir Arnold never cited references which he did not read (in talks he would even place doubts on proofs which he did not look at, even when belonging to major mathematician), and this is likely reflected in works of his school.

    Second, this is not very healthy attitude. Going toward references which are neither directly used nor highly recommended confuses the reader, who will consequently not know where to focus. On the other hand, it is easy to expand going recursively in other cited papers and other networked works (and the well known work will come up most frequently). Information lost within nonessential information is of course having smaller weight and by the rules of information theory it is in fact amounting to losing some information. As far as “embarassing” I do not buy it as an argument at all. The papers are written for readers and not for petting author’s vanity. Finally, if there are major misses the referees will almost certainly notice and object.

    • CommentRowNumber12.
    • CommentAuthorMark C. Wilson
    • CommentTimeOct 28th 2012

    Let’s start another thread on this one - it is interesting and has major implications in my opinion.