This.

You’d be surprised how much of a place’s physical infrastructure depends on a physical line. Automated fire alerts for high rises, security alarms, remote access for gates and doors, backup phone connections. A lot of this still uses old physical lines because it is easy to fix and highly reliable.

Now consider the infrastructure needed for specialized services like EMS, police, secure and classified buildings, federal agencies, embassies, smart traffic signals. Shutting down a network like that has massive implications for anything in society that relies on it, which is well beyond your cell phone plan.

$350k includes the salary but also all of the health insurance benefits, taxes, stock options, office space and perks, compute hardware, software services… the works. An employer will have an averaged overhead factor for their skilled workforce, which can be anywhere between 1.5 and 2.5 typically. A worker with an annual pre-tax salary of $140k could cost Google $350k in overall expenses per year. Labor is expensive.

Also, these people weren’t just making simple Python scripts. Most of them were contributing core functionality into Python itself and managing the internal Python version and the ecosystem of Google software stacks that depend on it.