Hacker News
points by jmyeet 4 days ago | 0 comments
The other way to look at this is that the standards committee was correct in rejecting this proposal because it added too much complexity without sufficient benefit. If so, hasn't the committee provided value?

I'm not arguing one way or the other (for C++ or Rust) but it's certainly worth considering.

I believe AI is the final nail in the coffin for C/C++ although it'll be a long, slow decline. C++'s backward compatibility with C cannot be resolved and you can never really be sure about memory ownership let alone safety. C++ in particularly is simply too complex. I didn't trust people to write correct C++ before AI without severe restrictions (eg Google's guidelines). I certainly don't now.

[–] Sharlin 4 days ago
(The Committee approved contracts as part of the C++26 DIS (Draft International Standard); it was the NBs (National Bodies) that didn’t. Because the Committee is made of NB representatives, the NB vote is usually just a formality. The situation is similar to a parliamentary committee spending many years drafting a major bill that’s ultimately rejected by the full parliament even though everybody assumed it would have the required (super)majority support.)
[–] swiftcoder 4 days ago
> The other way to look at this is that the standards committee was correct in rejecting this proposal because it added too much complexity without sufficient benefit. If so, hasn't the committee provided value?

The problem appears to be that they didn't have the majority needed to reject it at any of the previous points in the process, so we've now ended up with a stalemate where neither the Ayes nor the Nays can actually execute their position effectively

[–] vrighter 2 days ago
eh in rust you could just say it's a different edition, and claim that totally incompatible codebases are the same language
Hacker News