Experience in old technologies is a huge hindrance to progress - likely the biggest there is because it prevents people from understanding this new thing - it looks superficially like this old thing and so why not just stick with that?
This doesn't mean that old people are automatically worthless, so long as they play around with new technologies too - but looking at the number of people at my workplace who do not code outside of work, statistically the old guy sitting in front of you has spent the last ten years writing the same badly designed Java.
If you have spent the last ten years coding java and nothing but java what is the chance that you will choose Go to write that highly concurrent non-blocking network server that is going to be a huge difference in whether the business succeeds or falls? After all Java has nio so that should be good enough, right?
The thing is, those "brand new things" are usually not that new & more about fancy marketing than truly new stuff. Any developer with 15+ years of experience will have an easier time learning this stuff because of his experience & deeper knowledge on how it actually works.
It is entirely possible that the Java solution is more than good enough and will actually perform better. Experience teaches the value of benchmarks over hype and arrogant assumptions.
> Experience in old technologies is a huge hindrance to progress - likely the biggest there is because it prevents people from understanding this new thing
I don't think that's true; I think the closest thing that is true to that is that exclusive experience with a particular technology or closely-related family of technologies is a barrier to understanding novel approaches in the same domain that can't fit that same model. However, long experience with a variety of technologies in a domain, including continual learning of new ones, makes it easier to understand the next new one, even if it doesn't fit the model of older ones, because you are used to doing just that.
Further, that experience can help you from repeating the things that didn't work in the past with the new technology -- and help you understand where things that were tried in the past but failed might work in the parameters of the new technology.
> If you have spent the last ten years coding java and nothing but java what is the chance that you will choose Go to write that highly concurrent non-blocking network server that is going to be a huge difference in whether the business succeeds or falls?
Sure, that's a concern -- if you've spent the last ten years coding Java and nothing but.
OTOH, if you've got 30 years of experience that includes Pascal, C, C++, Java, C#, Objective-C, Ruby, Python, and Erlang -- plus some lesser-known, proprietary languages -- on platforms that include DOS, a variety of proprietary Unixes, NextStep, Windows, and Linux, you've probably got a better basis for evaluating the strengths and weaknesses of platform and language choices, and implementation approaches -- including considering new platforms, languages, and techniques in that evaluation -- for new projects than someone with shorter and less varied experience.
Just because programming isn't a real organized profession with continuing education standards doesn't mean that there aren't plenty of people that have been in it for decades that are serious about their field and that have spent a career doing more than just honing skill with a single, narrow technology.
