The other day, I don't know if it was a lightning strike or what, but I had a power outage for about a minute.

Of course, my PCs were also instantly blacked out.

Of course, all the files being created, etc., were wiped out, but still, when I could confirm that the entire system had rebooted.

I folded my hands and muttered 'Thank you, God' to the sky.

-----

I am not running a server or DB, so purchasing a UPS (uninterruptible power supply) seems like a bit much.

Well, one server has been working continuously for the last ten years.

The "Ebata Family Monitoring System" was also supposed to have been damaged.

When I shut it down, it would not usually restart, so I repaired things manually.

This monitoring server had one Wifi port available, so we also created an access point for experimental use.

-----

Around 1990, a theory was that programmers retire at age 30 (true).

This theory was a worldview in which a programmer's job was grueling -- overtime, all-nighters, load work, thin pay, and overwork -- and, indeed, it was like that.

If you ask whether the situation has changed now, it has not.

A program is a real pain in the ass because it is affected by the development environment, machine specs, and other software packages, regardless of the programmer's ability, and "when it doesn't work, it doesn't work no matter what you do. It is better to solve the problem by getting down on our knees and investing money and human resources.

Nevertheless, using OSS and better development environments has become much more manageable compared to the past.

And with generative AI as a partner, I feel that 'the retirement age for programmers has virtually disappeared.

If I ask the AI to create the first code in Japanese, it will generate the code for me. Once I understand the code created by the AI, I can modify it as I wish.

-----

Recently, I have been using it in a rather odd way.

For example, when I ask ChatGPT, "What is a multinomial logit model?

"Used in statistics and machine learning," "Extensions to logistic regression," "Extensions of logistic regression..."

However, this message doesn't make sense to me.

But when I ask ChatGPT to "try to make a simple program in the Go language for the multinomial logit model,

(Omitted)

It can make the codes like that.

I can get a quick overview of the "multinomial logit model" by reading this program (no need to run the program).

It is as if the time required to understand the mathematical model is accelerated by about 10.

Technical books are too dull to read.

-----

Well, I know that not many people use it this way. Still, in my case, with the help of "Generative AI + Programming Language," I hope to manage for the foreseeable future, even after I retire.

Though I still work in programming, I'm unsure if enough young people have programming skills.

There is a time limit until dementia sets in, but "hope" is "hope.