I agree with most of this but lots of people will go a whole career without massive changes. I do desktop dev and the tech is not very far from where it was when I joined the project 14 years ago. I still change files I changed in 2003.
If we found a good developer it wouldn't really matter much whether their skills was from 2007 or 2016.
It's not "tech" or "programming" that evolves quickly, it's just a few specific domains (web, mobile, ...).
Lots of others (desktop, embedded, ...) don't have that kind of churn at all.
True. My point is that if you've worked 14 years on a project, I'd like to know what you've done in those 14 years. Did you upgrade your toolchain, did you upgrade your main language version (eg. java 6 to java 8), did you consider webdevelopment and if so why didn't you change (I'm not pushing for web here, I'm just curious why you didn't follow the trend/hype). And what kind of substantial features/improvements did you create in those 14 years? Or did you only modify existing screens based on business specs?
Java didn't really evolve in the desktop development environment. We would still create Swing applications.
Yes, of course, sure. I mean it's a completely different world then and now (Multi core cpus, High dpi displays, language versions gone from C#1 to C#7 etc). Complete window toolkits swapped out etc. Feature set has probably grown 1000% (E.g. like Word 1.0 to 20XX).
So it's almost unrecognizable but the thing is that someone who was in on the project 10 years ago would be up to speed fairly quickly even though the language jumped 6 major versions and the window toolkit changed etc. And someone who wrote a swing app in '97 could jump on board this C#7 project and be productive immediately. The fundamentals don't change much.
When I look away from Javascript for 5 years I feel like I woke up on another planet.
> did you consider webdevelopment and if so why didn't you change (I'm not pushing for web here, I'm just curious why you didn't follow the trend/hype).
Same reason other heavy desktop apps and games aren't web based. Need the power of the local machine. It's not a db+forms app, it's a CAD+FEM package. There are of course some examples of "cloud cad" now but I doubt it will become mainstream any time soon.
Probably what is the core of the matter here is that the fundamentals of the application isn't the code that makes application tick, it's the domain logic. So while desktop moves slower than web/mobile, apps that are 90% e.g. physics etc moves slower than apps that are thin on logic and deep in presentation such as many web apps.
I suppose if our app was an AI app, then it would be pretty out of date because the domain moved so much in 10 years. But for CAD/FEM it doesn't.
If we found a good developer it wouldn't really matter much whether their skills was from 2007 or 2016.
It's not "tech" or "programming" that evolves quickly, it's just a few specific domains (web, mobile, ...).
Lots of others (desktop, embedded, ...) don't have that kind of churn at all.