I am writing some software that uses methods described in a few papers. The paper describes algorithms in terms of mathematical functions and equations. An example is, for instance, this one). The paper itself doesn't contain any source code - that is my own creation.

Where can licensing terms for the papers be found? What rights are implied when there is no reference to a license? What do I need to be aware of as I convert the algorithms described in the paper into a software package?

  • 1
    I don't see any source code in that article that you linked to. Are you taking their mathematical formulas and translating them to source code? Or am I missing some kind of source in that paper? – Thomas Owens Mar 21 '16 at 20:06
  • 2
    Taking their formulas. When I wrote methods I didn't mean functions in code, I meant way of doing things. – Jean-Bernard Pellerin Mar 21 '16 at 20:08
  • Do the papers you're using claim patent protection? Copyright protection wouldn't apply, since you're not copying source code from the paper. So probably no license is needed. – phoog Mar 21 '16 at 22:43
  • Anything you do can infringe on anyone's patent. That's just a fact of life. You have a research paper, but someone completely independent might have already patented what the research paper describes. If you are worried about patents, search for patents. Or don't, because having searched for patents can increase your liability. – gnasher729 Mar 22 '16 at 4:03
  • @gnasher729 To be clear, willful blindness / deliberate indifference are not defences to infringement or inducement. Also see Halo v Pulse scotusblog.com/wp-content/uploads/2015/10/… which held that for willful infringement, plaintiff needs to show that defendant was aware of an "objectively high likelihood that its actions constituted infringement of a valid patent". If you act on a reasonable belief of a patent's invalidity and that belief is not objectively baseless, the Fed Cir has ruled that is not willful infringement. (Pending USSC decision.) – user3851 Mar 22 '16 at 14:18

Copyright doesn't protect methods, only particular fixed expressions. 17 USC 102

Some methods (but not algorithms) may be protected by patent. Diamond v. Diehr 450 U.S. 175 (1981)

More exactly, "an algorithm, or mathematical formula, is like a law of nature, which cannot be the subject of a patent". However, in Diamond,

the respondents here do not seek to patent a mathematical formula. Instead, they seek patent protection for a process of curing synthetic rubber. Their process admittedly employs a well-known mathematical equation, but they do not seek to preempt the use of that equation. Rather, they seek only to foreclose from others the use of that equation in conjunction with all of the other steps in their claimed process.

Said another way: the algorithm cannot be protected, but if you are using the algorithm as part of a method or process that as a whole is patented, you would be infringing the patent. The paper does not have to disclose the patent - you could email the authors to see if they have any patent that protects any particular methods using the algorithm in the paper, but that doesn't rule out patents that the paper author is unaware of.

| improve this answer | |
  • So if the paper claims no patent protection, the algorithm described in the paper can be used freely, without any need to worry about licensing? – phoog Mar 21 '16 at 22:44
  • 1
    @phoog not necessarily. The patent might not have been disclosed in the paper. You could email the authors directly to check to see if they have any patent that protects any particular methods using the algorithm in the paper. Although, that doesn't rule out patents that the paper author is unaware of. – user3851 Mar 21 '16 at 22:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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