I've been a developer for over 30 years and there's _always_ churn. You will _always_ have to learn new things. That's just the way it is. Some things are truly better (C#) and some things are not (Flash). It often takes time to weed out the truly bad things, especially if the bad thing is the only thing that solves a noteworthy problem (how to provide high-end interactive animated content through a browser aka Flash). We've managed to move most of the basic concepts of Flash to the browser via HTML5 and that is very clearly a better solution.

The Angular 2 issue is somewhat curious. My biggest question isn't about the new teaser specification. It's what happens to the current version. It's open source, right? I suspect a number of corporate babysitters will provide the support required to make sure 1.x lasts for a decade or more. If you choose to adopt 2.0 for _new_ projects, so be it. Although why on earth the A2 architects chose to break XHTML standards baffles me. An attribute in parens? Seriously?

I personally see the new JS world as a step forward. It's absolutely a molten spew of new and unproven code, but it's also easier to build things than ever before. There are a ton of problems solved with a front-end MVC connecting to a service layer. A TON.

The hard part is that we need better tools all around for JS that we've been spoiled with in C#, C++, Java, and even Assembler.

But one more thing that's hinted at, but I'm suddenly asking myself. What if Anders Hejlsberg (C#/TypeScript) were to publish a monolithic front-end MVC/MVVM framework and its use was built into Visual Studio 2015, which is going to be free at the Pro level? That would throw even more F bombs into the JS world.

In any case, if you're a senior developer or architect and making decisions about technology, you better know what you're doing. You can pick 2% usage code, but you better own it.