3 ms·
Disclosure: I'm a just plain scientist, without the data. ;-) Since "data science" appeared on my radar (in other words, on HN), I've noticed that we fling the
by analog31 6y ago
Disclosure: I'm a just plain scientist, without the data. ;-)
Since "data science" appeared on my radar (in other words, on HN), I've noticed that we fling the "scientist" and "engineer" terms around without asking whether the practitioners have science or engineering backgrounds, or something else such statistics, math, programming, etc.
It strikes me that "do we need scientists or engineers" is not unique to data science/engineering. I think we need both, but of course it's an open question as to how many of each are needed. Also, both "scientist" and "engineer" are loosely defined in practice, with some overlap.
Painting with a broad brush, a scientist wants to learn how things work. An engineer wants to make things that work. If you're in the business of making things that work in order to sell them, then you need lots of engineers, but maybe a few scientists. It's about 10:1 at my workplace.
Overlap? Of course. Engineers use scientific knowledge and methodology. Scientists have to make things work in order to make experiments and theoretical computations work. Also, scientists often show up in emerging fields before they become recognizable engineering disciplines, for many reasons: 1) We need the newest stuff, right away. 2) We are opportunists by nature and necessity. For instance many of the older "software engineers" at my workplace have science degrees, but the younger ones all have computer science degrees. The oldest people I know who did programming, digital logic, and embedded systems, did not have degrees in those areas. The youngest ones all do. At my college, the physics professors had personal computers, and were ripping them apart, while the CS professors used the mainframe. That's part of why I chose to major in physics, though I was interested in programming.
Another reason for overlap is that both "scientist" and "engineer" titles include things that are not strictly within either area. A lot of people with science degrees end up working as technicians, marketeers, salesmen, managers, etc. A lot of people with engineering degrees do little or no quantitative engineering, but work as programmers, designers, marketeers, salesmen, managers, etc.
Something to ponder is if there are differences between how scientists and engineers think and approach problems. Naturally there's a lot of folklore about that, probably little hard evidence. That's where I think you should start if you want to know whether to hire scientists, engineers, or both.