> I've just explained that sand-boxing causes issues with file access, clipboard sharing etc.

You've explained that flatpak has issues with file access and clipboard sharing. My iphone does sandboxing too, but the clipboard works just fine on my phone.

I don't think "failing clipboards" is a problem specific to sandboxing. I think its a problem specific to flatpak. (And maybe X11 and so on.)

> If you remove control, you remove people's freedom.

Sandboxing gives users more control. Not less. Even if they use that control to turn off sandboxing, they still have more freedom because they get to decide if sandboxing is enabled or disabled.

Maybe you're trying to say that security often comes with the tradeoff of accessibility? I think thats true! Security often makes things less convenient - for example, password prompts, confirmation dialogue boxes, and so on. But I think the sweet spot for inconvenience is somewhere around the iphone. On the desktop, I want to get asked the first time a program tries to mess with the data of another program. Most programs shouldn't be allowed to do that by default.

> Pretending you can make some theoretical system where those trade off don't exists just isn't realistic.

I think you might be arguing with a strawman. I totally agree with you. I don't think a perfect system exists either. Of course there are tradeoffs - especially at the limit.

But there's still often ways to make things better than they are today. For example, before rust existed, lots of people said you had to make a tradeoff between memory safety and performance. Well, rust showed that by making a really complex language & compiler, you could have memory safety and great performance at the same time. SeL4 shows you can have a high performance microkernel based OS. V8 shows you can have decent performance in a dynamically typed language like JS.

Those are the improvements I'm interested in. Give me capabilities and sandboxing. A lot more security in exchange for maybe a little inconvenience? I'd take that deal.

> You've explained that flatpak has issues with file access and clipboard sharing. My iphone does sandboxing too, but the clipboard works just fine on my phone.

> I don't think "failing clipboards" is a problem specific to sandboxing. I think its a problem specific to flatpak. (And maybe X11 and so on.)

There are other examples.

e.g. There are other things that become a PITA on the phone. Want to share pictures between apps without them having full access to the everything. You need to manually share each picture between apps.

The point being made is that it causes usability issues. What those usability issues are will vary depending on platform. However they will exist.

> Sandboxing gives users more control. Not less. Even if they use that control to turn off sandboxing, they still have more freedom because they get to decide if sandboxing is enabled or disabled.

Anything that gets in my way is something that taken control away from me. Unfortunately giving me full control comes with dangers. That is a trade off.

> Maybe you're trying to say that security often comes with the tradeoff of accessibility? I think thats true! Security often makes things less convenient - for example, password prompts, confirmation dialogue boxes, and so on. But I think the sweet spot for inconvenience is somewhere around the iphone.

No usability and control.

BTW, Your sweet spot is a platform which is the most locked down.

> On the desktop, I want to get asked the first time a program tries to mess with the data of another program. Most programs shouldn't be allowed to do that by default.

Well I don't want to be asked. I find it annoying. I assume that this is the case when I install the program. So I don't install software in the first place that I think might be risky. If I need to install something that I might think is iffy then I find a way to mitigate it.

> But there's still often ways to make things better than they are today. For example, before rust existed, lots of people said you had to make a tradeoff between memory safety and performance. Well, rust showed that by making a really complex language & compiler, you could have memory safety and great performance at the same time.

You aren't selling it to me. I got so annoyed by Rust that I didn't complete the tutorial book. Other than the strange decisions. One thing I hate doing is fighting with the compiler. That has a cost associated with it.

I spend a lot of time fighting with the TypeScript compiler (JS ecosystem is a mess) as a result to have some things work with TypeScript you need to faff with tsconfig and transpilers. Then once you are past that you have to keep the compiler happy. Frequently you are forced to write stupid code to keep the compiler happy. That again has a *cost*.

> V8 shows you can have decent performance in a dynamically typed language like JS.

I work with JavaScript a lot. While performance is better, it isn't actually that good.

There was also two secondary effects.

- Websites ballooned up in size. Also application development moved to the browser. This meant you can lock people in your SaaS offering. Which reduces control/freedom.

- There is a lot of software that is now written in JavaScript that really shouldn't be. Discord / Slack are two of the slowest and memory hogging programs on my computer. Both using Electron.

> Those are the improvements I'm interested in. Give me capabilities and sandboxing. A lot more security in exchange for maybe a little inconvenience? I'd take that deal.

Again. It is a trade-off that you are willing to take. I am willing to make the opposite trade-off.

You seem to be arguing that adding complexity reduces freedom, but I don't think that's true in a reasonable interpretation of the word.

Your argument would suggest that virtual memory takes away user freedom, because it's now much harder to access hardware or share data between programs, but that sounds ridiculous from a modern perspective. I think it's better to keep freedom and complexity separate, and speak about loss of freedom only when something becomes practically impossible, not just a bit more complex.

Yes, you do:

> Anything that gets in my way is something that taken control away from me. Unfortunately giving me full control comes with dangers. That is a trade off.

No I am not. The example given was ridiculous and absurd and you are doing exactly the same thing.

There is a big difference between basic memory protections and what was being discussed.

This is the issue with a lot of people that work in software. They take the most ridiculous interpretation because "that is technically" correct while not bothering to try to understand what was said.

The problem is that if what "really counts" is too vaguely defined, then it's hard to pin down and argue the point.

Virtual memory probably isn't what you meant, but take something like user privilege separation. It's usually considered a good idea to not run software as root. To interpret the statement generously, privilege separation does restrict immediate freedom: you have to escalate whenever you want to do system-level changes. But I think josephg's statement:

> Sandboxing gives users more control. Not less. Even if they use that control to turn off sandboxing, they still have more freedom because they get to decide if sandboxing is enabled or disabled.

can be directly transposed to user privilege separation. While it's true that escalating to root is more of a hassle than just running everything as root, in another sense it does provide more control because the user can run arbitrary code without being afraid that it will nuke their OS; and more freedom because you could always just run everything as root anyway.

Maybe josephg's sense of freedom and control is what you're saying there is a trade-off between. But the case of privilege separation shows that some trade-offs are such that they provide a lot of security for only a little bit of inconvenience, and that's a trade-off most people are willing to make.

Sometimes the trade-off may seem unacceptable because OS or software support isn't there yet. Like Vista's constant UAC annoyances in the case of privilege separation/escalation. But that doesn't mean that the fundamental idea of privilege levels is bad or that it must necessarily trade off too much convenience for control.

I think that's also what josephg is suggesting about sandboxing. He says that the clipboard problem could probably be fixed; then you say, "but there are other examples". What remains to be shown is whether the examples are inherent to sandboxing and must degrade a capabilities/sandbox approach to a level where the trade-off is unacceptable to most.

> The problem is that if what "really counts" is too vaguely defined, then it's hard to pin down and argue the point.

It really wasn't. It isn't hard to understand what was meant.

> Virtual memory probably isn't what you meant,

No it wasn't and there is no need to put "probably". It was obvious it wasn't.

> can be directly transposed to user privilege separation. While it's true that escalating to root is more of a hassle than just running everything as root, in another sense it does provide more control because the user can run arbitrary code without being afraid that it will nuke their OS; and more freedom because you could always just run everything as root anyway.

The difference is that there are very few things I need to run as user directly daily as root on my Desktop Linux box. I can't think of anything.

However having to cut and paste a meme into ~/Downloads so I can share it on Discord or Slack is a constant PITA. If you sandbox apps you have to restrict what they can access. There is no way around this. The iPhone works the same way BTW. I know I used to own one. You either have to say "Discord can have access to this file", or you have to give it all the access.

> Maybe josephg's sense of freedom and control is what you're saying there is a trade-off between. But the case of privilege separation shows that some trade-offs are such that they provide a lot of security for only a little bit of inconvenience, and that's a trade-off most people are willing to make.

No they are a false sense of security with a lot of inconvenience. The inconvenience is inherent and always will be because you will need to restrict resources using a bunch of rules.

> Sometimes the trade-off may seem unacceptable because OS or software support isn't there yet. Like Vista's constant UAC annoyances in the case of privilege separation/escalation. But that doesn't mean that the fundamental idea of privilege levels is bad or that it must necessarily trade off too much convenience for control.

There are many things that seem like they are fundamentally sound ideas on the face of it. However there are always secondary effects that happen. e.g. Often people just ignore the prompts, this is called "prompt fatigue". I've literally seen people do it on streams.

Operating systems are now quite a lot more secure than they were. So instead of going for the OS, most bad actors will use a combination of social engineering to gain initial entry to the system. The OS security often isn't the problem. Most operating systems have either app stores, some active threat management.

If you are running things from npm/PyPI/github without doing some due diligence, that is on you. This is well past what non-savvy user is likely to do.

> I think that's also what josephg is suggesting about sandboxing. He says that the clipboard problem could probably be fixed; then you say, "but there are other examples". What remains to be shown is whether the examples are inherent to sandboxing and must degrade a capabilities/sandbox approach to a level where the trade-off is unacceptable to most.

It is inherent. It obvious it is. If you want to share stuff between applications like data, which is something you want to do almost all the time. You will need to give it access at least to your file-system. The more of this you do, you will either have to give more access or having to faff moving stuff around. So either you work with a frustrating system (like I have to do at work), or you disable it.

So what happens is you only have "all or nothing".

> If you want to share stuff between applications like data, […]. You will need to give it access at least to your file-system. The more of this you do, you will either have to give more access or having to faff moving stuff around.

Why are those the only answers?

If we had free rein to redesign our computers from the ground up, there’s lots of other ways that problem could be solved.

One obvious example is to make copy+paste be an OS level shortcut so apps can’t access the clipboard without the user invoking that chord. Then just copy paste stuff between applications.

Another idea: right now when I invoke a shell script, I say “foo blah.txt”. The argument is passed as a string and I have to trust that the program will open the file I asked - and not look instead at my ssh private keys. Instead of that, my shell program could have access to the filesystem and open the file on behalf of the script. Then the script can be invoked and passed the file descriptor as input. That way, the script doesn’t need access to the rest of my filesystem.

If we’re a little bit creative, there’s probably all sorts of ways to solve these problems. The biggest problem in my mind is that Unix has ossified. It seems that nobody can be bothered making desktop Linux more secure. A pity.

Maybe it’s time to give qubes a try.

> However having to cut and paste a meme into ~/Downloads so I can share it on Discord or Slack is a constant PITA.

Why round trip it through the file system or Files.app? That seems like extra (annoying) work On my iPhone, I copy the meme onto the clipboard and then I open discord/slack/signal/Whatsapp and find the right channel/chat, and paste right in there.

> It isn't hard to understand what was meant.

At least two independent people understood you in the same way. So just dismissing it isn't productive.

> PITA. If you sandbox apps you have to restrict what they can access. There is no way around this.

This has nothing to do with freedom though.

> You will need to give it access at least to your file-system.

On Qubes, you copy-paste with ctrl+shift+v/c and nothing is shared unless you actively do it yourself. It becomes a habit very quickly (my daily driver). Sharing files is a bit harder (you send them from VM to VM), but it's not as hard as you want it to look.

> At least two independent people understood you in the same way. So just dismissing it isn't productive.

Two people that we are aware of.

BTW, I often encounter this when talking to other techies. People go to the most ridiculous extremes to be contrarian. Often they don't even know they are doing. I know because I used to engage in this behaviour.

So I feel like I am well withing my rights to dismiss it.

I didn't say you weren't within your rights. I said it's counter-productive for the discussion.

I think it is counter productive to bring up ridiculous examples, which was obviously not what I meant.

Both things can be counterproductive simultaneously.

> Any security mechanism has a weakness or it will be bypassed by other means. So all this will give you a false sense of security.

> The moment you think you are safe. Is when you are most unsafe.

This is demonstrably false. Qubes OS has the lowest number of CVEs, even less than that of Xen. Last VM escape in it was found in 2006 by the Qubes founder (it's called "Blue Pill").

Also: https://news.ycombinator.com/item?id=27897975

You are only thinking of attacking computer directly itself. Often people socially engineer access to a computer system. Many UK super markets were hacked, using some of the software that is very secure, because people managed to socially engineer access.

There is nothing and I mean nothing that is completely secure.

> There is nothing and I mean nothing that is completely secure.

You're not wrong, but dismissing security because there are always other threats is just security nihilism. See my link.