As a software engineer, we are always eager to get effective tools to boost our productivity.
And in our daily work, we are often unsatisfactory with the existing tools and would like to have better ways such as better GDB script config, Vim script and some Python script to make boring things automatic.
However, it is actually a trade-off since making tools also needs time and energy. It does not give productivity boosting immediately. Therefore, how do you judge whether it is time to stop work and to make some tools to ease your future pain?
Best Answer
I "make tools" when one of these is true:
The "risk" for the 2nd option doesn't have to be huge - the cost of building one small tool is usually small, so if all you save is the risk of running a 10-minute build again once a week, it will usually repay itself very fast.
I try to make tools as small as possible - just make the task a little bit easier now, and maybe improve again next time.
This means you only fix the biggest pain each time, and not creating fixes to problems that don't really hurt you.