I think it's like this: what used to be group in regular permissions output is a union of group and ACL mask in ACL output. Mask sets the upper limit of what ACL can do, so if mask is rw- then it's impossible to set a r-x ACL permission because allowing execution is not allowed.
This seems to be more geared towards enterprise environment where it could be complicated to cleanly define groups, since you can only give access to one you might run into a problem where dept. A needs access to that directory but also person G from dept. B and person K from dept. C.
My inner child is dead and I killed them in about second year of university because of fear of other people and lack of money for anything beyond food and tuition -> no fun times except for the computer -> depression, unmedicated for 3 years.
And now that I have money, I'm a mess after years of neglect, and I forgot how to have fun, and I still suffer from depression.
But as good as that sounds in theory, I would rather keep my memories untampered. Brains are really bad at actually remembering things as they were at the time when they were remembered, any tampering might as well go unnoticed. I would rather not experience my favorite media for the first time again and also not risk getting my entire personality rewritten because of a bug or even worse - deliberate action.
Hey, I'm not blaming all Japanese in America for Pearl Harbor, I'm blaming all Japanese in Japan for Pearl Harbor.
If you cannot express your opinion under the threat of death (but more likely, if you are a normal person and not a politician, a slap on the wrist or in worst case jail time), does your opinion really matter?
I have nothing to say about Iraq invasion except that one invasion for dumb reasons and shitty intelligence make all invasions for dumb reasons justified I guess.
They had anywhere from 115 months to 19 months, depending on where you start counting, to make up their minds about who's at fault in this war. If this much time isn't enough...
Futureproofing, I guess, but also I'm sure there are things other than language models that can benefit from a dedicated processing unit, like photo processing, smaller models, Lens, etc.
If it's all in the cloud - yes. If some/most of it runs locally - still yes with a caveat - hardware acceleration can make it run both faster and at a lower battery cost, same how every computer has a dedicated graphics chip, whether as a separate expansion card or a separate module on the CPU. Yes, you technically can do all that math on CPU, it's called software rendering, but rendering done with a purpose-built chip is so much better. Same logic applies to "AI".
Keed using it, set up
smartdto check up on it regularly and send you an email if something goes wrong.