Post #2429451
2026-04-30 04:33 UTC
@argv_minus_one@mastodon.sdf.org Singularity & Midori.
Smalltalk and Lisp systems could count too.
Spectre vulnerability. Not sure about the details, but the hardware would also need to lack speculative execution,
There are ways around that and different ways to do very similar that don't run into the same problem.
and nobody's going to accept such a severe performance hit.
I would.
Those computational models enable efficiency gains that probably largely outweigh the loss of prediction anyway (as several layers of indirection are obviated by a smarter base API/ABI & capabilities).
Now, comparing such Language+Capability systems with prediction and without probably would favor those with, but comparing those LC systems without prediction and highly-indirected systems (Linux-likes) with prediction is a different matter.
Replies (1)
-
@argv_minus_one@mastodon.sdf.org 2026-04-30 04:43
@lispi314@udongein.xyz That's a very extraordinary claim. It requires very extraordinary proof. If there were a way to achieve performance on par with what we have now, yet do so without the security risk and extra circuitry required for speculative execution (and the security/stability risk of ambient-authority monolithic kernels), wouldn't we be using it already?