I, Chris Doty-Humphrey, release all of my work on PractRand to the public domain. All statistical tests in PractRand are wholly my own work, and are public domain. All the auxillary code, including the command line tools, are my own work and public domain. All of the recommended RNGs are unencumbered. I do not know of any patents covering any part of them: jsf* was designed by Robert Jenkins, and he has released it to the public domain. The implementation is mostly his code with a little of my own. isaac* was designed by Robert Jenkins, and he has released it to the public domain. The implementation is mostly his code with a little of my own. sfc* was designed and coded by myself, and is public domain. efiix* was designed and coded by myself, and is public domain. hc256: From the eSTREAM project page: "The stream ciphers HC-128 and HC-256 are not patented and are royalty free. Anyone can use HC-128 and HC-256 free of charge." I started with the reference implementation and modified it slightly. mt19937 is an unencumbered algorithm. The implementation included is my own, and is public domain. trivium: From the eSTREAM project page: "To the author's knowledge Trivium is not covered by any patent. Trivium is also free available for any use." The implementation included is my own. xsm* was written by myself, incorporating some ideas from svfuerst on usenet comp.lang.asm.x86, and is public domain. salsa: From the eSTREAM project page: "My policy is that Salsa20 is free for everyone to use. I am not aware of any patents or patent applications covering Salsa20. I will update the ECRYPT Stream Cipher project when necessary." see also: http://cr.yp.to/snuffle/ip.pdf The implementation included is my own. chacha: I believe the author intended this to be covered under the same license as salsa, but unfortunately I can't find any explicit statement to that effect. I think he hasn't gotten around to it yet. If the IP situation is important to you, use Salsa instead just in case. They have identical interfaces and features so that should be no issue. The implementation included is my own. The "NotRecommended" / "other" RNGs included in the full version... are more numerous and less important - they are intended for comparison purposes, and not intended to actually ship with any real product. The code for them was either written by myself or copied from a publicly accessible web page, forum or usenet post, or academic paper. I do not know of any patent situation involving the algorithms, but have not paid any great attention to them.