FreeBSD - it won’t be easy, but I’ve been a BSD guy at heart for decades… You will learn a lot and eventually be able to create better systems, but it will be years before you should risk putting anything important on a system - as a noob you have a lot to learn the hard way. Once you think you know FreeBSD you should try the other BSDs, and things like gentoo linux: you will really learn how this works.

You can follow the advice of the others and get a system going sooner. It isn’t a wrong choice, but you won’t learn as much and if something doesn’t work the way you want you are stuck since you can’t dare change anything. As such I have to advice against it despite all the time/effort my advice will cost you.

You can start with used but modern x86 - the n100 line - has very low power usage and will long term be a better investment. A pi is about the same cost once you get the accessories needed and uses as much power to get work done, but can do less work. (If the computers are idel the pi wins)

Not when the target is expected to know that. Spelling it out is for when the reader might not know.

I had found that. A lot of projects are early releases that have not been touched in years - is that because they are stable or because the author gave up before making them useful? Which is why I want not a list but an opinion from someone else doing this.

You will spend more maintaining those older cars than buying a newish used car. Depending on what age of course, but things wear out and eventually they become hard/expensive to fix.

Thanks for the reminder, I need to do that again.

They exist, but hard to find. Most people doing this are hobbyists doing the conversion for their own fun. They might help you with your project but for liability reasons they are just helping.

A V8 has a lot of low RPM torque so if you drive with a light foot it can do very well. Most of the energy from an engine is used moving the car, not engine losses. A smaller engine is always going to do better at the same load, but the difference isn’t going to be large if everything else is equal (which it often isn’t)

Most people who are cheap enough to do without the amenities still want them and so will buy a used car with all the amenities over a new car at the same price without.

It is less valid for accounting software than cars on the road. If you mess up accounting software CEOs go to prison for tax fraud. If you mess up car software it is someone else who dies, but no prison time for the CEO.

Points don’t fail that often, and I just need to file them every few months.

If you can get the app. I looked at Waydroid and other apps, but discovered that app I needed was only on the google play store. (I want to see when my kid’s bus will arrive - there is no web site, just an app. I suspect it needs google logins or something to use, but I gave up before I figured out how to download the .apk)

I have enough of a machinist background to doubt the threads are anywhere close to perfect. However if you are saying more than good enough I will agree.

Or is it better to save a few bucks now and save it for next year when something new comes out that is faster anyway. Maybe there is a new codec that matters in 3 years but nothing today supports: so either way you are forced to replace your server.

There is no right answer, you are taking your chances when planning for the future. There are many computers more than 10 years old still working just fine in the world, and it is possible that whatever you buy today will be as well. We get enough press releases that we can predict what will happen next year close enough, but in 5 years we have much less information. There is no way to know if saving money is a good choice today or not. I can come up with scenarios either way.

Look at power use. Often last generation hardware uses more power for the things you do today and so the few dollars you save today are made up with in the power bill over the next couple years. (though if you use that new hardware to do something the old couldn’t do the new will use more power!)

If there is only a few dollars difference in price go for the best. However when there are hundreds or even thousands of dollars it becomes a harder decision.

Patents have a short life span. The patent wall keeps expiring and then everyone can use it no cost. the big improvements are long gone and all they can patent is small improvements you can do without (though you may not want to). They also run into those making things do their own r&d and have their own patents.