planish

That could work fine, probably? Or you could use it on the same machine as other stuff.

I did use older Android, and I agree that the new permission model is absolutely much better for the use case of running apps that you do not trust or even like. I can scan a coupon with the camera today without having to worry that the store's app is going to be taking pictures of me tomorrow.

But that's hardly any of what I use my phone for. So I pay a lot of the costs of more hoops to jump through to allow stuff I actually want, while not really getting much of the benefit of being able to use malicious applications relatively safely.

And the one time I had a real permission problem, it was Snapchat trying to bully me into giving it access to all my files so it could "detect screenshots" before it would let me talk to my friends. And Android permissions were no help there, because the app can still tell if I reject its requests and won't get booted from the store for refusing to work until I grant access to everything, even though I do not want to.

The whole system seems to me to be designed to make people feel like their privacy is being protected, by popping up all the time to say that unused permissions have been removed and hey look at all these privacy options you have. It does indeed stop people from spying on your location and camera all the time without you noticing. But while the little permanent green dot is flashing every five minutes when your location is sent to Home Assistant like you explicitly asked, and you are trying to decide if you want to let Zoom use Bluetooth headsets just right now or on an ongoing basis, Google is hoping you don't notice that the OS and most of the apps are designed to extract value from you rather than to serve your interests.

It's now safer to run the evil apps, but they're still there trying to do evil.

I think that one was the cows burp methane, which is a greenhouse gas. So if you apportion the greenhouse gas emissions over the delicious hamburgers, you make more climate change by making a cow burger than a veggie burger. So we should cease the production of cows as part of our attempt to not make our planet terrible. And buying cow burgers to eat is contrary to the goal of ceasing cow production.

What's the security problem with Compatibility Mode? Is it just that it lets you let an app run with more permissions than it otherwise has on the new APIs? Or does it turn off a bunch of mitigations?

The Android permissions churn seems meant to protect people from applications: previously you could just say you need GPS, install, and then use GPS all the time. But untrustworthy apps started tracking people all the time, so Google declared that now only Google Maps is allowed to track people all the time, and that everybody else has to do a new location access ritual. If I have an old app that I trust (or wrote!) but doesn't do the ritual, I ought to be able to convey to the OS that I trust the application anyway. The machine works for me, not for Google's idea of what my privacy preferences are.

I don't see how a developer not implementing new permissions models is the developer not caring about security. I guess a more robustly sandboxed app is more secure than a less robustly sandboxed app? But just because a security enhancement like that is available doesn't mean it's actually worth doing, and the user experience of the new system (get sent to settings to toggle on file system access for a file manager) is often worse than before.

Having new development is better for the user than not; they will get features and improvements. But having to do development to prevent the user from losing features over time is a pure cost to the developer. The rate at which it currently happens makes it unnecessary hard to do projects that aren't shaped like commercial subscription services.

ZFS zRAID is pretty good for this I think. You hook up the drives from one "pool" to a new machine, and ZFS can detect them and see that they constitute a pool and import them.

I think it still stores some internal references to which drives are in the pool, but if you add the drives from the by-ID directory when making the pool it ought to be using stable IDs at least across Linux machines.

There's also always Git Annex for managing redundancy at the file level instead of inside the filesystem.

Actually I think this might be https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1023748 and you might be able to get around it by having an older Java installed or set as the main one when installing the CA certificates package.

Your root problem is:

dpkg: error processing package ca-certificates-java (--configure):
 installed ca-certificates-java package post-installation script subprocess returned error exit status 1

It can't finish installing ca-certificates-java because the script that is supposed to set it up isn't working. So it also can't finish installing anything that depends on it, since it doesn't want to run their scripts until ca-certificates-java is installed properly.

Maybe uninstall ca-certificates-java and what depends on it, and then reinstall it alone, and see if you can get a message about why exactly its script is failing?

Yes?

You can also shout into the void yourself. Eventually somebody might notice and reply.

You can watch the local or federated timelines and if someone says something interesting you can follow them and reply. Or you if you get linked from elsewhere to a good post you can follow the author. Or you can see a list of popular hashtags (or look up a hashtag) and see posts in them, or post to them.

What else is there supposed to be to do there?

You can use interest rates to convert between stocks and flows of money. If the prevailing interest rate is 5%, a thing will produce 5%, or 1/20th, of its actual value every year. So you can take the annual cost of something and multiply by 20 (and vigorously wave your hands at compounding) to get its actual value.

A $10/month subscription costs $120/year, or $2,400 over 20 years. So it's equivalent to a $2,400 purchase.

You can also think of it as, you need to set aside $2,400 in investments to pay for your subscription, e.g. in retirement. Or, if you ditched your subscription you could afford to borrow $2,400 more to e.g. buy a house. Or, you as a customer are the same value to the business as $2,400 in capital, minus whatever they have to spend to make the thing.

You should think a lot about a $2,400 purchase.

Games are a good example. One might want to publish a game and then work on the next game, not go back to the first game again and add dynamic permission prompts for the accelerometer or recompile with the new SDK or whatever. But someone also might want to play Space Grocer I before Space Grocer II-X to get the whole story.

The fewer breaking changes there are, the lower the burden of an app being "supported" is. Someone might be willing to recompile the app every couple years, or add a new required argument to a function call, but not really able to commit to re-architecting the program to deal with completely new and now-mandatory concepts.

Even on software I actively work on that is "supported" by me, I struggle with the frequency of e.g. angry messages demanding I upgrade to new and incompatible versions of Node modules. Time spent porting to new and incompatible versions of a framework is time not spent keeping the app worth using.

If you write a commercial program and sell it once, you are probably not going to be selling new copies in 10 years. If you keep getting paid you should indeed keep working. But if you stop working on it, it is better for the finished software to last longer.

Windows 11 has a "compatibility mode" that goes back to before XP. Android has a dialog that says that an old APK "needs to be updated", regardless of the continued existence of the original developer or whether the user is happy with the features and level of support.

It is this attitude of "we don't need to think about backward compatibility because we are powerful and all software has a developer on call to deal with our breaking changes" that causes software to go obsolete very quickly now. User needs also change over time, but not nearly as fast.

Not all of the same weaknesses. If it's just "let the judge move stuff around because they're a judge", then yeah. But if you implement any sort of security on it, you can say that the judge can only move stuff when also countersigned by the jury, who were demonstrably selected by a fair random draw, or something.

And even if you don't do that you still have a great record of which judge exactly is stealing everyone's stuff.

You can't just wave a blockchain wand and get a government that works, but you can just wave a blockchain wand and get an accountable record of things.