Why software has degraded in the past two decades

A revolutionary development is only really worth aspiring for if it improves the existing situation. I don't think there's an issue with (and I certainly don't think it's a sign of weakness) relying on tried and tested methods, processes, and/or technologies if they work for the thing you need them to do. Sure, it's a good thing to constantly improve. But inventing something completely revolutionary isn't inherently positive - it's only positive if it serves a purpose...

too many bad programmers out there. there's software, then there's software as an art. The demand is just so much that they don't care about the quality as much.

That said, still lots of improvements went on as well.

I think software has degraded because the companies that make it are more focused on making money than they are on making good products.

When I was a kid, my mom used to buy the same brand of bread at the grocery store. But then one day, she went back and tried to buy that brand of bread again—and it was completely different! It wasn't even the same shape! I remember her complaining about how she couldn't find the bread she was used to buying, and eventually she just gave up and bought something else.

I think what happened is that some company bought out the maker of my mom's favorite bread, and then changed it so they could sell more of it. They wanted to make more money, not make better bread.

That's why I think software has degraded in the past two decades—because companies are focused on making money instead of making good products.

Oof, I opened the link, and his page background nearly blinded me!

This rant is a bit forced and doesn't really prove the claim. He actually makes three separate claims in a fairly short piece, and doesn't properly support any of them.

There is one thing I absolutely agree with: whatever benefits we've gained from the last 40-odd years of software abstraction, "fewer bugs" hasn't been one of them. If anything, our programs are more brittle and bug-prone than ever before. I have some background in automated software verification, and while there are some great technologies in that space, they're not about to come to our rescue anytime soon.

I think the only way our industry can reliably produce higher-quality products is to make far fewer of them. This is the case in markets that are strictly regulated. The software driving our aircraft and our banks is much less buggy that e.g. Apple's music player, but it is also far more expensive to run, maintain, or add capabilities to.