> [Blog-author:] The professor thinks that every student needs to learn how he drives his horse-and-buggy. Maybe a few do. But he shouldn’t be angry that most of them prefer to use a car.

I'm siding with the professor here. If the goal of education classes was to churn out the right answer as fast as possible, we'd just let all the humans go home and have an automated system copy the answer-key.

There's a huge difference between:

1. I understand this concept and the way it applies, even if I'm not going to regularly exercise it to this same degree of abstraction. Here is a solution which is evidence that I have gone on the intended mental journey.

2. I asked the mad-libs machine and it told me this was the solution, and it seems to work, gimme grade plz.

> [Blog-author:] It seems to me that for a long time, well before ChatGPT, you could develop software without knowing elementary computer science.

You've always been able to paint without a Renaissance Portraiture class, but that doesn't justify cheating in the class by having someone else do your homework, and if you don't think the curriculum is valuable, maybe don't sign up for it.

People who try to prove an argument by analogy forget that analogies are analogues not abstractions. Computer science is not a horse and buggy to F1 racing car AI, computer science is the abtraction of design principles which allowed buggies to exist, allowed horse vetinarians to exist, and allowed engineers to make engines to replace horses.

It's a sub-arc branch of mathematics. Do we need mathematicians to have horses and buggies? No. We have mathemanticians to work out how many horses and buggies we need and how many we can afford, and to count their legs and divide by 4 for an approximate answer.