3 ms·
I feel like writing almost the opposite: An Engineer's Regret: Focusing Too Much on Math. I did a lot of calculus, some functional analysis, numerical methods,
by BeetleB 7y ago
I feel like writing almost the opposite: An Engineer's Regret: Focusing Too Much on Math.
I did a lot of calculus, some functional analysis, numerical methods, etc. I've never needed even basic calculus for my engineering job. I did use it a few times, but mostly because I was looking for an excuse to use it. No one cared, and demonstrating such skills played no role in my annual review. Furthermore, everyone around you has forgotten this stuff, so the system will shift to not valuing math.
Of course, jobs do exist that need heavy math. They are not the norm, and there is a lot more competition to get those jobs. With the exception of machine learning, your employer will value you as a SW engineer more than a math wiz, precisely because the demand for the former is much higher compared to the supply.
(I honestly don't regret it - but the value of knowing advanced math in the professional world is overestimated).
- YeGoblynQueenne 7y agoThe thing is, software engineer salaries are going down and data scientist salaries are going up and the data scientist jobs are taken by people from the sciences, who know that kind of math, that CS courses maybe are not that heavy on. So there is a great big incentive to know "maths" (continuous maths). And if one does not know, the incentive is for one to learn.
- selestify 7y agoInteresting, can you give a pointer to the data on software engineering salaries declining?
- BeetleB 7y agoIt seems you agree with me: >With the exception of machine learning,
- codesushi42 7y agoI don't know if what you are saying is true. Where is the data? But I do believe in the long run, you will be right. Software is saturated, and a lot of the infra has been built or commoditized. The value will be there for people who can analyze data. Too bad I don't find analyzing data as fun or as interesting as banging out a programming problem.
- leonidasv 7y agoI think both are held together. If you have a lot of data, but your underlying system is simple, you don't need a lot of people to analyse it. Your demand for new people crunching that data will come as the complexity of your software grows. Think of it this way: having 10 endpoints and 10TB of plain-text data vs 1000 endpoints and 1TB of plain-text data. The later will surely require more time to be analysed, even though the amount of data is smaller.
- codesushi42 7y agoDisagree. Software is saturated and consolidated; the infrastructure has been built and is owned by large companies. And that trend will continue. Data on the other hand is still growing, and there is enough already that has not been analyzed. Traditional software engineering will still be important, but no longer glamorous. The future belongs to the data analysts.