I started to program very early - it was the most entertaining thing that I had going at 13 years old (well that and reading SF of course). And I honed my skills and learned many things very early.

As an example, still very vivid in my mind, my TurboC compiler broke at some point and suddenly started refusing to compile a source file. It was a large file (too large: at the time I have not yet formed a good taste on what’s “just right” source file size for me) with a lot of macros use. So I did a binary search on it (or “Thanosed” it) by halving the file and compiling and then identifying the broken half, and then the half of the half, etc. This thought me three things:

• Don’t always trust the tools (sometimes it is zebras)
• Split the problem into smaller bits
• Do not respect computers

This last is probably the most important thing I learned early on. I respect the sea - it is not entirely predictable and it can kill me (or worse: it can kill my kids so I thought them to enjoy the sea but also respect it from a very early age). I respect my fellow engineers (even when I warn against excesses). But computers, at least those that are not attached to X-ray machines, in general do not kill or maim. They can of course be connected, again, to something dangerous or risky: your bank, trading account, nuclear launch codes, etc. But in programming there is nothing like that - you are just wrestling with the problem and you should have no respect for machine executing your code itself. Do not approach it with trepidation but confidence that you can make it do what you want.

Last modified on 2025-03-14