There used to be claims that are literally impossible as per the whole field of CS, that were later silently removed. That’s not a good look.

So they are young and foolish but motivated and talented.

I can live with that.

It makes the project look less professional, or legitimate, but that's about it.

Imho features are non-features until they work. As described. Reliably.

I believe you that as a user you don't care, but for the community that implements the features, they exist in the imagination of the devs, design, planning and early implementation attempts for a long time before you can use them.

You're both right and wrong.

Right in the sense that it can take a really long time to release a work in progress feature.

Wrong that some of the claims made initially were phantasmagorical and also debunked very quickly.

Also, vaporware is a a thing. I'm perfectly willing to stay by the claim that GNU Hurd will be useless to me until the day I die.

For reference GNU Hurd has been in "dev, design, planning and early implementation attempts" for 33 years, and I imagine I'll live 50 more years (fingers crossed!).

It feels like an enthusiast project, where the developers didn't know some things are hard or impossible, and they claimed their language will solve them (or that it already solves them). Looks like they've learned something now :)