Suppose you implement an algorithm published in a research paper and use it in your startup. Is it necessary to somehow "cite" the paper? Is it as informal as dropping the author(s) an e-mail and saying, "Hey, cool algorithm. Just wanted to let you know I used it..." or is there some formal legal process?
You don't need to cite the paper. You only need to cite your sources if you are publishing further academic literature. The whole point of publishing literature is to enable other people to actually use the findings in practical life.<p>If it so moves you, you can write a nice blog post about the algorithm and send an email thanking the author.<p>You should also be careful. The original author or their institution may have patented the algorithm.