THE MOST COMMON STUPID QUESTION I GET FROM SOFTWARE TEAMS: Attempting to find a

THE MOST COMMON STUPID QUESTION I GET FROM SOFTWARE TEAMS:

Attempting to find a scientific or axiomatic rule for what is an economic problem.

All programming is approximately the same. The question is only the performance demanded, cost of production, cost of maintenance, and cost of change.

If you break software into responsibilities and understand the difference between resident(executables), and instances(node.js), temporary(scripts), and transactional (database) then from that point it’s all economics.

In general, a software developer wants to reduce his costs, unless he has only his self-interests in mind and not the results of his actions.

Never hire anyone who says they want to stay on top of tech. People stay on top of tech if they’re good at producing results.

Those questions that are not economic – and there aren’t many of them – are empirical. And empirical tests defeat theoretical rules in all cases at all times.

The rules in software are the lifetime of execution, responsibility, design patterns in UI, middle, and db, and costs. That’s it.

Risk used to be significant in the past. The major risk today is not the software itself which has become relatively cheap, but the business model we anticipate with the software, and the success in providing transaction cost of use that justifies the cost of the use.

Curt Doolittle