On software longevity

tags

Engineering

Sometimes I wish I could go back and see where all of my old code has gone. Surely some of it has been discarded.

Some is likely in Github limbo, sitting in a repository with an ancient last commit timestamp.

After a decade in the industry, though, there must be teams out there who have the displeasure of working with code I wrote years ago.

I’ve ideated with a mental model for thinking about the short, medium, and long-term scales of aging software; 3 days, 3 months, and 3 years.

Why 3? Mainly because it feels uncomfortable, but within reason

3 days feels close. What happens to the software 3 days from now is almost instantly impactful to me.

3 months feels out of reach individually, but a business lives by the quarter. New teammates will join, pritorities will shift, and projects will grow.

Will the code I’m writing, or the process I’m introducing today be more helpful or more of a roadblock when those events occur?

3 years may as well be 100 years. Technologies will change, my framework choices will be outdated, and I might not even be on the same project.

The same question applies.