On Holden's seven guidelines for scientific computing and development of open-source community software

Two decades ago, Helge Holden proposed seven guidelines to improve the way new achievements and results in scientific computing were presented, evaluated, and compared in contemporary scientific literature. In this essay, written as a tribute to Helge on his 60th birthday, I revisit the guidelines and point out why they are still valid today seen from my perspective, working as a contract researcher at the interface between mathematics and applications in industry. Developing new computational methods usually involves a lot of experimental programming. Over the past decade, my research group has developed an open-source community code that today has hundreds of users worldwide. I discuss some considerations that have gone into this development and present a few lessons learned. Moreover, based on this experience, as well as from development of professional software for our clients, I present advice on how you can be more productive in your experimental programming and increase the impact of your scientific results. Science is what we understand well enough to explain to a computer. Art is everything else we do. - Donald Knuth, Foreword to the book A=B (1996)