I think there is a lot of fake innovation happening these days. I bet if the programmers of the 70s had had the same amount of computing resources available as we have now they would have built great systems even with the tools available then. Other than the cloud and more computing and network power I don't think software development has really moved forward much.
Here's a nice infographic: https://medium.freecodecamp.org/the-biggest-codebases-in-his.... Unix 1.0 (1971) has a third of the number of lines of code as the average iPhone app. I know LOC is a very poor proxy for complexity, but I have a hard time believing that 70s engineers could replicate the functionality of Google's 2 billion LOC with tools like ed.
If as a system is small enough for a single programmer to hold in their head, you can remove a lot of layers and be maybe 10x more productive. Those '70s systems were all built like that. But eventually you reach the limits where a system is too big for that and you have to take the hit and build it collaboratively; it's also the only way to get systems that are maintainable in the long term.
> Other than the cloud and more computing and network power I don't think software development has really moved forward much.
So a person from the 70s who was instantly transported to today would feel like the only difference is that instead of having our own servers we ship things to Heroku? Please...
Have you seen the Alto demos? Quake would be new to them, maybe the browser. As for shipping to Heroku IBM was renting mainframe timeshares in the 1960's.
