5 ms·
Ask HN: How has internet changed software development?
- new_guy 5y agoI think it's made a lot of people overestimate their competence and overvalue their 'skill'. They think being able to Google/SO something is the same as knowing it. Give them a laptop without internet access and most 'developers' wouldn't even be able to get it setup nevermind write 'hello world' from scratch.
- mooreds 5y agoHaha, I'd put it slightly differently. It has changed what skills are important to software development. It used to be knowing things cold or being able to look them up in a book was critical (having the right book being a precursor). Now, it is knowing how to search, knowing what terms might be used, and knowing who to trust that is more important for many types of work. In some ways, it's like the transition between oral history and written history. Both were ways of storing knowledge, but one was far far easier for more folks to access. Of course, as a dev, it's also important to be able to experiment and test, and the internet has changed that by making computing power easier to access (whether heroku, cloud providers, or even just things like requestbin).
- oceanghost 5y agoPre-internet: Most programmers I knew had walls of books of reference material. Books like "PC Interrupts", the "386 system design guide", "Zen of Code Optimization" etc. And I didn't know many programmers. Computers were expensive and so was the software and reference books. I was lucky enough to be able to afford these things. Also, everything was just slower before the internet. At the rate things change now-- it doesn't make sense to publish books about most programming topics that change much faster than than a book could possibly be published. The internet also changed computing in general because, that's when computers started working against us (the beginning of surveillance capitalism).
- tuckerpo 5y agoThere's too much information. Any jamoke with an internet connection can write a blog and consider themselves an epic hax0r for following <latest nifty development principle>. People get bogged down worrying about idioms - am I following SOLID? Would Uncle Bob like my code? Am I writing unit tests? Should I mock my foo class, or maybe make it a stub? Should I integrate my build with Jenkins, and set up email automation when builds fail? Should I follow the factory decorator observer enterprise pattern? The result is lots of time spent sweating over whether or not you're following the latest snake oil pamphlet of "best practices" published on whatever hip tech site you frequent. The days of putting your head down and cranking out functional code and worrying about how "clean" it is later are mostly gone. Programmer productivity has plummeted. Complexity has increased exponentially for simple tasks. It's not great.