There's nothing even preventing the second form from working either. Just put the right shebang at the top of the script and it'll run through that interpreter. I've been on fish for a decade, but still write all my shell scripts in Bash. It's never been an issue.
Quick note on #2 - there aren't really any issues with storing your encryption root passphrase in a file. If the file is owned by root, with no read permissions for any account, only root can access it. Since it's stored on an encrypted dataset, and your initramfs is as well, it's unreadable when the machine is off. Lastly, if anybody _does_ have a root shell on your machine, they can change the encryption passphrase without needing to know the current value.
In short, I'm not sure there are any real issues with having it on disk but unreadable by anybody but root.
In general I agree with you but there is one difference - a sneaky user with physical access can read it and _not_ change it, vs changing it. The latter is more detectable. But this is minor.
Yeah. Unfortunately, ZFS encryption is missing a few creature comforts of something like LUKS. I've stuck with native OpenZFS mechanisms, though, to keep the complexity sprawl to a minimum.
Absolutely - I know that but thanks for pointing that out again. There is no real "use case" for NOT storing the key into a root owned file. However, as I don't do it for myself there is no way of accidentally deleting the file, copying it quickly from my system to another drive when I accidentally left a root shell open and went to the restrooms (that never happens;) and the one single place I store the key (my head) is pretty much unreadable for everyone except me (at least for now :-) Being paranoid doesn't mean they are not after you :p
Since I reboot my notebook only about once in a month it is no real hassle to enter the key twice 12 times a year :-)
I booted a G4 Mac the other day, running 10.4.something. I was thrown back in time to a period where OS X was clearly their flagship software stack. Everything was coherent and cohesive - and shockingly - fast. I'd daily 10.4 again if it could operate on the modern internet comfortably!
Nevertheless, I also remember that in the 10.4 days, OS X had the reputation of being sluggish compared to Linux or even Windows (I guess it was Windows 7 at the time?). And it kinda was. Bouncing ball when launching an app.
Snapdragon support is decent to great these days, and importantly it's all in the mainline kernel tree.
Edit: though it should be said that what I think is good might be a far cry from you think is good. I did use a Thinkpad X13s as my primary work machine for 6 months, though.
Looks like the post was neutered, but this is another source. The speed of the car was well in excess of what the road conditions could tolerate. Horrible crash, but the fault can be placed squarely on the driver of the car.
Don't forget the V series in there. I have an Intel(R) Core(TM) Ultra 7 258V in my Thinkpad. I think they're still being made. I bought an open box Thinkpad T14s Gen 6 with it - they come with a nicer GPU than the Ultra 7 255U.
The V series is a one-off thing Intel did, but they don't have a direct successor planned.
Previously, they had a P series of mobile parts in between the U and H series (Alder Lake and Raptor Lake). Before that, they had a different naming scheme for the U series equivalents (Ice Lake and Tiger Lake). Before that, they had a Y series for even lower power than U series.
So they mix up their branding and segmentation strategy to some extent with almost every generation, but the broad strokes of their segmentation have been reasonably consistent over the past decade.
Very interesting. I was a bit out of the loop on Intel mobile CPUs; I looked up the benchmark specs for it when purchasing and saw that it generally trounces the 255U.
I've been really quite happy with it - most of the time the CPU runs at about 30 deg C, so the fan is entirely off. General workloads (KDE, Vivaldi, Thunderbird, Konsole) puts it at about 5.5 watts of power draw.
reply