Okay science aficionados and practitioners let me ask you a simple question: Who’s more important, the telescope maker or the telescope user? Is it the person who discovers the astronomical body or the person who enables the discovery?
Let me take a radical stance: they are equally important.
Okay I’m playing games here, most people would probably agree with this statement. So what’s the big deal? Well the big deal is that culturally we tend to celebrate the telescope users. For example, how many people know the names Hans Lippershey, Zacharias Jansen, or Jacob Metius (three people associated with the discovery of the telescope)? Compare this with the single name Galileo Galilei, credited with the discovery of four of Jupiter’s moons, sunspots and the phases of Venus.
In all fairness Galileo contributed an enormous amount to science, was elevated in name during tumultuous times (asked to recant his theories under pressure from authorities), and has the sort of name that fits nicely into a popular musical or song. He also refined telescope design (i.e., became a telescope maker). By comparison the telescope makers were simple craftsman who “were by and large illiterate and therefore historically often invisible.“
If it were simply a cultural artifact I wouldn’t be writing this blog. Unfortunately, I believe that this bias towards telescope users goes beyond cultural perception, it affects contemporary technology decision makers too. I am also of the opinion this is negatively impacting the transition of technology to benefit the public good, especially in this current age of technological complexity.
One way that this comes about is that in the minds of too many technology leaders and policy makers there is lazy distinction made between “science” and “engineering” (i.e., telescope users and makers). Such leaders think about funding “science” and often give short shrift to “engineering” (e.g., implementation details). I believe this attitude is becoming increasingly harmful, more than ever insights into the workings of our physical world depend on outstanding telescope makers.
To be more concrete, we experience this phenomenon at Kitware in many of our collaborations with world-reknowned research programs. Too often our esteemed telescope user friends are hamstrung by inadequate software, poor data management resources, and lack of computation expertise on staff. Fundamentally the balance is wrong: science is in desparate need for better engineering, yet the engineering is starved because of “lack of novelty” as judged by a review committee. In the worst case the end result can be failure or inconsequential conclusions, but even in the best case when great discoveries and data are found, there is often no transition path, meaning that the impact on the public good is small.
I can say for sure that in the computational sciences this situation is only getting worse; the data is getting bigger, resolutions higher, the algorithms and data more complex, and required computer skills more demanding. And since the computational sciences are critical to almost every scientific endeavor we engage in these days, we need to rethink the balance between science and engineering. For example at Kitware we carefully balance our staff between those that are engaging in research versus those building the research infrastructure. While many of our staff can hold their own with any researcher anywhere, we take equal pride in our engineering prowess. As a matter of speculation, I believe that our researchers make more of an impact because of the excellence of the telescope makers supporting them.
I think we all recognize the need for telescope makers and users, and that both are equally important to the ongoing success of technology. However, to effectively engage in science we need to rethink the balance between engineering and science, particularly in our support for scientific computing.