I think that if you put a flag it effectively means removing those features from Scala, because it will declare them as dangerous. If that happens, then you have lost to Kotlin / Ceylon because these languages have an edge on Scala in terms of ecosystem (IDE / hibernate). The only chance Scala has is in allowing more sophisticated language constructs.
I think the source of the problem, at least in this blog post, is that extension methods are limited today. They are done with implicit conversion which is limited for obvious reasons I guess (not allowing type A to arbitrarily become B). I think a feature focusing just on adding extension methods is required, even if it makes the language spec larger.
The point of the original article is that there's no point in avoiding blocking on IO while allowing blocking on CPU.
Furthermore, since node.js is single threaded, what's wrong with blocking on IO in that single thread? The process hangs, is put to sleep by the OS and wakes when there's IO available. You gain a simpler model of programming than using callbacks/continuations
I think the source of the problem, at least in this blog post, is that extension methods are limited today. They are done with implicit conversion which is limited for obvious reasons I guess (not allowing type A to arbitrarily become B). I think a feature focusing just on adding extension methods is required, even if it makes the language spec larger.