So, the question is: 'Does a thing a smart guy told us about computer systems in 1967 holds true now, 59 years later, and does it apply to my particular technology'? :)

React is not representative of how Facebook's whole tech stack works. It is a small composable piece of tech they created, then open sourced, then it was shaped by many external forces, and then was adopted by thousands other big and small companies and individuals.

Also, you are not Facebook, but you are also not Basecamp/37signals.

I have written a lot of VanillaJS and a lot of self made framework-like code, plus I've used the one off creations of other developers. React is far superior.

Balderdash. A lot of smart people got together and built something that makes getting to a working product faster for you than if you had done it all yourself. This is the core of what makes us human and allows us to advance as a species. This kind of attitude would have us still living in the dark ages with no electricity or running water. How about be pragmatic and use what is out there, learn from it, and contribute back to the body of knowledge so we all get that much smarter? 10 years from now ReactJS may or may not just be a distant memory, but the legacy of development and thought that it has spawned will live on in whatever next thing comes out, and that is the true sign of something worthwhile.

Stop worrying about the arrangement of source code behind your apps and build things that delight, engage, and change the world.

What happened to "we stand on the shoulders of giants"?

His presentation is, expectedly, slighted toward an increasingly outdated tech stack and I hear a note of jealousy in the dogmatism.

If it works best for you and your product then use it.

I would say this is a classic "I have an opinion" topic :D .... you could as well say "you should not use libraries" if you not know what they do :)

I think in some cases react slowed my process donw a lot because I had to adapt to the new concept .... now I'm "okay" with it but not good. And I've talked with a friend of mine who actually does mainly Javascript and loves react. he said that react people tend to over-complicate solutions for problems. (We were talking about react-router vs a simple history object callback)

But I'm full-stack polyglot with strong tendencies to the system-architecture, backend and devops :) I cannot compare myself with a real frontend-developer who gets to take care of this all the time and has practiced it for a long time.

I think that's an interesting question it's expectation vs reality... many humans think we are so smart, we forget how long it takes to comprehend and master a new system and often we don't know how to handle it.

And there is a lot about religious taste ofc... :)

Sam Stephenson strikes me as the kind of guy who says things without thinking.

React is quite an non-opinionated view library that makes no prior assumptions about what you are using it for. While Facebook might have built it with the original intent of making UI composition a lot easier for their developers and UI teams, it's not a niche library. It is no more a "Facebook specific" technology than PHP is. Look at other solutions, Rails came out of 37signals, Django was created by a newspaper.

Some of the best frameworks and libraries were spun out of existing organisations/projects built to meet company needs. And look no further than Angular which came about from Google's internal Greentea team who work on a private CRM tool that Google use.

The bottom line is: it shouldn't matter who created a particular framework, tool or library. You should always pick a particular piece of technology because it meets your needs.