Back in school, a teacher went up on stage during assembly time to give a speech, a passionate plea. She didn't do that very often, but this time she had something important to say.

Turned out she was the one overseeing the all important computer class. At the time, classes on computers were extracurricular. There was one dedicated room in the whole school for it, mostly running 286x CPUs.

She was up the stage because she was upset that hardly anyone signed up for the classes. You have to pay extra to join, you see. And you stay behind after school hours to do it, after which you have to figure your own way home because school buses have left. That school wasn't known to produce state-class talent, so I wasn't surprised about the lackluster reception.

This teacher went on to give an interesting thesis, if I could call it that, about how computers will be very important in the future (meaning now). According to her, computers will run everything. The skills we learned for searching library indexes will hardly be applicable anymore, we have to learn computers to do that instead. Bottom line is if we kids didn't learn to use computers, we will fall so very far behind.

So I signed up for classes. I'm not sure if it was because I heeded her call or I thought it was a good idea. It's likely that I got FOMO'd enough by this teacher.

As soon as the first class started, I felt something was off. The machines were MS DOS (not even Windows 3.1x), which all of us had to learn to navigate. I guess that is fine, but I already had a sense that this is not quality shell is supposed to look like.

The class isn't about teaching MS DOS though, which might have been quite acceptable. We navigated MS DOS in order to get the thing that's being taught: WordStar. It was a textual user-interface word processor that preceded or competed with WordPerfect, all before a GUI Microsoft Word came around and gobbled the entire word processor business.

I recently learned that it was the software that George R.R. Martin wrote all his novels in, because WordStar was what he started with and couldn't move away from.

I had this sinking feeling in every class session. We were not told what we should use WordStar for. But we were taught how to ctrl-this and ctrl-that to perform tasks we didn't know why we needed to do. I was thinking "am I supposed to memorize all these? How am I going to do that if I barely had one-hour access to computers each week to practice? Or am I supposed to memorize them like we memorize history books?"

Anyway, why WordStar not WordPerfect? Nobody asked, none of that was discussed.

Barely a few years later, Windows 95 came out, the computer world was changed, everything we learned about WordStar was good for nothing.

The teacher was right. Computers indeed ended up running the world, learn it we should have. The vision and spirit were in the right direction, the time and money spent executing these computer classes gave us less than zero leg up.

It's hard to know where to begin with this categorical error. First, it's not a computer class as much as it's a word processor class. They either thought word processor will stay the way was and won't get easier to use; or they thought everybody will need to use word processor because it's just better than a typewriter.

Even at the time everybody had a sense that hardware and software advanced quickly. Yet they didn't foresee that tools like WordStar would improve to the point where we don't even need workshops to learn to use MS Word.

They could've taught emacs or vim and we've still end up net positive now. They didn't have a sense of what's Lindy.

If things expire fast, the only sensible thing to learn would be things that will always be true. They are usually referred to as fundamentals.

But tools are easy to learn and straightforward to teach. It's easier to sell "ChatGPT workshops" than to sell "machine learning 101". The gratification for students are immediate: use the tool for season, get results, make profit.

The barrier for teaching and learning fundamentals are usually higher. Understanding something often requires understanding something else, turtle all the way down. Against such cost, benefits are hard to justify if you can't produce immediate results.

Picking what to learn is similar to investing. Investments that are easy to make deliver very little return. Things that are easy to learn don't stay relevant very long.