Sometimes I may come up with an equation in mind, so I want to search for the related material. It may be the case that I learn it before but forget the name, or, there is no name for the equation yet. In this case, I may be able to recall a reference book. Searching in Internet can be a fast way though.

However, there are cases that I get some idea and equations (maybe a modified one) myself. So I want to know whether there are any other people working on it before. Because of the vagueness, it is difficult to search it by keywords because I do not know the 'name' of this idea: it is either too board or no results.

I can try to search google scholar, arxiv and maybe the prola, but there is no support of equation pattern matching. For example, entering ∇⋅V in google give you no useful results (it is better if you input the 'divergence of potential').

Is there any good way to search papers in this case? It is even better that I can use the combination of equations and keywords.

Edit: Another reason is that Mathematician should have done some deep analysis on the related mathematical topics. Yet it is difficult to know their results because there is a gap between the terms used in physics and mathematics. It would be really useful if we can find and learn their results.

  • 1
    $\begingroup$ It is a practical question and I guess it belongs here because most of us should have similar experience. If it is not appropriate, then close it. $\endgroup$
    – unsym
    Nov 28, 2010 at 17:59
  • $\begingroup$ Nice question. Honestly, I don't know how to do it either. I just try to associate keywords I think are relevant. And I just play around until I find what I need. Or not. But one thing is certain, the search can be interesting in itself. Sometimes, I wasn't able to find out what I needed, but I still harvested all kinds of interesting articles. $\endgroup$ Nov 28, 2010 at 18:09
  • $\begingroup$ Excellent question! I never realized but a search engine for mathematical formulas could be really useful! How many times have we all worked with formulas to shape them into a state where we can solve them? $\endgroup$ Nov 28, 2010 at 18:12
  • 1
    $\begingroup$ On the other hand, for integer sequences, there is a nice website cataloguing them: oeis.org . I wish someone would think up something like that for formula's. Maybe I should think about it. $\endgroup$ Nov 28, 2010 at 18:14
  • 2
    $\begingroup$ @Lagerbaer, not a bad suggestion, but the problem is that there are so many idiosyncratic notations that you might miss what you are looking for just because someone wants to write his derivatives in "Newtonian fluxions" style instead of Leibniz notation or whatever. And just the problem of different names for the relevant variables is daunting enough. $\endgroup$ Nov 28, 2010 at 18:59

2 Answers 2


You can perform $\LaTeX$ search - that is, write formula in LaTeX in an appropriate search engine:

However, as one can type the same expression in different ways and with different symbols, I never used it it practice. (Anyone did?)

  • $\begingroup$ 0 down vote I've never used it neeither, but trying the example (\nabla V) asked in the question seems to give relevant results : latexsearch.com/latexFacets.do?searchInput=\nabla%20V&stype=exact $\endgroup$ Nov 29, 2010 at 10:08
  • $\begingroup$ @Frédéric: Actually OP wants \nabla\cdot V :) $\endgroup$
    – kennytm
    Nov 29, 2010 at 10:19
  • $\begingroup$ I have tried it before, but it is not useful though $\endgroup$
    – unsym
    Nov 30, 2010 at 1:29
  • $\begingroup$ That is very cool. $\endgroup$
    – Matt Ball
    Dec 10, 2010 at 19:47
  • $\begingroup$ Handy link! I never knew of such a site. $\endgroup$
    – Noldorin
    Dec 10, 2010 at 22:09

There was a similar question at Mathoverflow. I think it contains some useful references and discussion so it's definitely worth checking out.

Usually I would post this as a comment under the question but probably nobody would notice it anymore. And seeing that the other answer is also quite short I hope this is fine.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.