That's an assertion, but not anything to back it up.
First, threads have been implemented as libraries many times. Second, if checks need to happen theg can happen at debug run time if they can't happen at compile time. I don't know what specifically has to be integrated into a language here that makes throwing away the enormous amount already built in other languages.
These points are all clearly addressed in the paper I referenced.
> That's an assertion, but not anything to back it up.
Yes it's an opinion on style, not a falsifiable claim.
> First, threads have been implemented as libraries many times.
The title isn't intended to be taken quite so literally. The author explains why they think these don't work correctly.
> Second, if checks need to happen theg can happen at debug run time if they can't happen at compile time.
But languages don't have mechanisms to implement these kind of checks.
> I don't know what specifically has to be integrated into a language here that makes throwing away the enormous amount already built in other languages.
You're mistaken. It's not 'integrated into', it's 'taken out'.
Libraries let you add things, but to design a model for parallelism you generally want to take things away. You want to take away the ability to do things outside the model's rules.
It's the same reason.
Your dataflow semantics need to be part of the language semantics, otherwise they're bound to be loosely defined and even more loosely enforced.