4 ms·

For people interested in science (finance would also work) there is an infinite number of things to be done to be useful (or simply interesting) at various leve

For people interested in science (finance would also work) there is an infinite number of things to be done to be useful (or simply interesting) at various levels. Doing calculations by writing functions, reading files, etc. is a great way to progress. One can take some equations from an intro physics/chem/bio book and get started. Or doing some budgeting/personal finance/investing simulations using personal banking CSVs or online stock data. It's a good alternative to game programming in that it may be more obviously pragmatic and appeals to people (like myself) who are thoroughly uninterested in games. I imagine it's a very different style of programming as well; less interactive and more geared towards producing results rather than an experience. For more experienced programmers looking for interesting and useful side projects (maybe to show off skills for a job search), I think there is a lot of opportunity to work with scientists to help them with their research. It's actually pretty easy to reach out to scientists by going to public talks and seeing what they do, then chatting them up and seeing how to make yourself useful. The hard part is of course sticking with it...

Elderofozone 10y ago
I did just that in my early career as a hydrogeologist. I wanted to impress my soon to be masters supervisor, so I programmed a simple analytical model that is used in my field. In doing so, I learnt about how cautious groundwater modellers have to be in interpretation and fundamental problems in the mathematics. My supervisor still brings it up everything I see him at conferences.
- cossatot 10y ago
  Nice! I learned Matlab to do nonlinear optimization of GPS data to solve for fault locking depths and slip rates, and then a year or two later learned Python to do run FEMs of thermochronology data (also to get fault slip rates) on Amazon's servers in parallel. (Matlab was not really up to that task.) None of that was to impress my advisor exactly, but it worked...

Elderofozone 10y ago

I did just that in my early career as a hydrogeologist. I wanted to impress my soon to be masters supervisor, so I programmed a simple analytical model that is used in my field. In doing so, I learnt about how cautious groundwater modellers have to be in interpretation and fundamental problems in the mathematics. My supervisor still brings it up everything I see him at conferences.

cossatot 10y ago
Nice! I learned Matlab to do nonlinear optimization of GPS data to solve for fault locking depths and slip rates, and then a year or two later learned Python to do run FEMs of thermochronology data (also to get fault slip rates) on Amazon's servers in parallel. (Matlab was not really up to that task.) None of that was to impress my advisor exactly, but it worked...