4 ms·
It's not really surprising. Engineering (which most programming is) is the economic application of scientific knowledge for human value/advantage. Thus that i
by xyzzy21 5y ago
It's not really surprising.
Engineering (which most programming is) is the economic application of scientific knowledge for human value/advantage.
Thus that implies several things:
• the core engineering is defined by economics, psychology, sociology, etc. at least 50% with technical/scientific knowledge/skills being the other 50%
• the money engineering generates is subject to human need and preference which is why the above liberal arts areas matter
• competitive value to the customer defines all projects, all hires and all salaries and decisions are only partially rational even with rational R&D, buying and selling processes
• factors that define the negotiation of value for price involve soft sciences which are messy but completely unavoidable. Those of us who "excel" in those areas have realized this and learned skills related.
• all of this is what "business" actually is!
• thus if you do software seriously, you also should be serious about business - this is analogous to Alan Kay's quote: "people who are serious about software should build their own hardware".
If you want "purity", become a scientist - nominally they don't care about anything but the purity/attainment of knowledge. But that alone does NOT pay well. That's why there are 200+ PhDs in physics waiting for each US tenured university position but companies doing engineering can't get enough and pay for that.
Science is only 100 years old as a profession. Engineering is many 1000s of years old because it has practical application which everyone else in society can easily agree to fund.