The joy of having a properly implemented capability system is that, well, you can create arbitrary capabilities.

You don't need to give a process/component the “unrestricted network access capability” -- you could give it a capability to eg “have https access to this (sub)domain only” where the process wouldn't be able to change stuff like SSL certificates.

EDIT: and to be clear, fuchsia implements capabilities very well. Like, apart from low-level stuff, all capabilities are created by normal processes/components. So all sorts of fine-grained accesses can be created without touching the kernel. Note that in fuchsia a process that creates/provides a capability has no control on where/to who that capability will be available -- that's up to the system configuration to decide.

Ok, give me access to a subdomain I control and I’m phoning home and there’s no way you can restrict mysubdomain.foo.com/phonehome vs mysubdomain.foo.com/normal - and even if you tried to do path restrictions, I can arbitrarily side-channel phoning home with normal access (which by the way you can’t unless you’re sniffing the encrypted HTTP session somehow).

Also imagine you are trying to run a browser. It’s implicitly going to be able to perform arbitrary network access and there’s no way you can restrict it from phoning home asides from trying to play whackamole blocking access to specific subdomains you think are it’s phone home servers.

That’s why I said “semantic” capabilities aren’t a thing and I’m not aware of anyone who’s managed to propose a workable system.

I imagine one could create a capability such that the app gets a way to shove bits in and a way to get bits out, but no knowledge of the IP address or anything like that. A phone (or set of phones) that are already connected and have no keypad.

> there’s no way you can restrict mysubdomain.foo.com/phonehome vs mysubdomain.foo.com/normal

Of course you can!

With capabilities you can tell a program: "if you want to communicate with the external world here's the only function you can use :

`void postToMySubDomainSlashWhatever(char* payload, size_t size)`

Ok great. Now I put the phone home stuff within payload. It’s a game of whackamole you’re bound to lose. Like I said - if I control both endpoints, it’s going to be very hard for you to simultaneously give me a pipe connecting them while controlling the set of messages I’m allowed to send.

Sure, like you said, having control of the endpoints you could communicate anything if you can transmit bits. That’s unavoidable.

But don’t lose the perspective on the benefits of such an architecture. Considering the networking access example:

* If your process gets compromised it won’t be able to access the attacker’s C&C server. It wouldn’t have access to any other stuff that the process didn’t already have for that matter.

* You wouldn’t be able to use http. It would be https only.

* Your process wouldn’t need a lib to talk HTTP. It would just need to talk the IPC protocol (whose wire-format and related details are standardized in Fuchsia which allows for the binding code for (de)serialization to be auto-generated).

* You wouldn’t be able (for better or worse) to mess with SSL certificates, proxies, DNS resolution, etc.

Consider another example -- file access. Say your app wants to access a photo. It doesn’t have access to the filesystem nor to the user’s folders -- it only has access to, say, an “app framework services” capability (eg. services from an UI-capable OS like Android) whose one of the “sub-capabilities” is requesting a photo. When your app does that request the ‘system’ opens a file selection GUI for the user to pick a photo. Note that the photo picker GUI is running on a different process and your app doesn’t know and can’t access anything about it. All that matters is that your app will receive an opened file handle in the end. The opened file handle is a capability as well. The file handle would be ready-only and wouldn't need to actually exist in any file system. In this example, before handing the file descriptor to your app, the “system” (or whatever process is implementing the ‘photo-picking’ capability) could process the image to remove metadata, blur faces, offer the user to edit the image (and maybe actually save it to a persistent store), log that access for reviewing later, etc.

(We already have something kinda similar in Android, but the implementation is not from first principles, so it’s very complex and prone to issues (requires an obviously non-POSIX Android userspace using lots of security features from the Linux kernel to sort of implement a microkernel/services like architecture)).

EDIT: adding the detail that Fuchsia's IPC lang ecosystem has autogen features due to its standardization.

I really don’t know what point you’re trying to make. I am 100% in favor of capabilities and think it leads to better decomposed software with better security boundaries (provided the software engineers put in the work to separate components across process boundaries and the APIs make it convenient to do so).

All I said was that capabilities don’t solve the spyware problem and they largely don’t. They help protect you write software that itself can’t be hijacked to become uncontrolled spyware due to a compromise but if I am selling you software with “malware” bundled you’re going to have a hard time isolating the functional and “malware” bits (malware here being defined as software against the users wishes and intents).

You’ve extolled the benefits of it and they’re great and I think I largely agree with all of that, but it’s completely irrelevant to my initial point that it’s not a silver bullet for the vendor intentionally bundling malware into the code they’re distributing.

In my experience lots of folks simply won't work with capability systems no matter how good the implementation is or whatever level of security and configuration granularity is provided.

For many people it's just extra friction in search of a use case.

It makes testing a lot easier honestly. Also keep in mind that mobile apps and web apps are fairly capability oriented these days, so I wouldn't say no one will work with it...

I'm just hearing about capability systems today, so your experience is undoubtedly richer than mine, but I'd estimate that we're just scratching the surface re: ways to harm somebody by making their tech behave in surprising ways.

Maybe once those harms are all grown up, we'll find that fancier handcuffs for our software is worth a bit more than "just extra friction."

I am curious what your experience is with capability based security? They are still incredibly niche(unfortunately) so I’ve never had a chance to work with one at a job.

Most components don't need to talk to the network though and therefore do not. The ones that do can do powerful things but creating narrower capabilities to restrict what they can do is very much feasible.

I’m not against capabilities. I’m just highlighting it’s for the developers to implement protections against malicious intrusions against the OS, not for users to protect against developers doing malicious things.

While there is no direct UX exposing this to an end user, it hypothetically be used as the basis of such UI. The parent of a component gets to ultimately decide what capabilities it routes to a child component. It's not like landlock where the process decides to sandbox itself after it's already running. Similar to a user constructing a VM to run a hypothetically malicious program, the same could be done much more lightweight with a fuchsia component.

All I said is that in the general case you’re not going to be able to rely on capabilities to do things like prevent phoning home or otherwise doing things you semantically define as harmful. This isn’t a UX issue - this is a technical issue. Capabilities, outside from very rare circumstances, can’t enforce it no matter how you structure this. The only rare circumstances is if you can restrict access to servers that aren’t owned by the same people as wrote the component. As soon as you give access that allows access, they can implement it in ways that capabilities can’t prevent.

Your perspective is coming from a very rigid all-or-nothing mentality and I don’t think it’s wise to see things that way.

Sure, a web browser that needs to open arbitrary network connections can be built to phone home. But nearly none of the components it’s built out of can. The image decoding and rendering libraries can’t touch the network, the rendering engine can’t touch the network, and nor can the dozens of other subcomponents it needs to work.

Your installed editor extensions can’t phone home even if the editor itself can. Or perhaps even the editor itself wouldn’t be able to, if extensions are installed out of band.

Your graphics driver vendor can’t phone home, your terminal can’t phone home, and on and on and on.

A solution doesn’t have to be perfect for it to be an improvement, so stop acting like it does.

But your editor extensions can’t phone home only if your editor sandboxes them into a separate process. Hint: VSCode doesn’t do such sandboxing and neither do most editors that I can think of.

Anyway, you’ve just proven my point with “install extensions out of band” - you’ve ceded that it’s a losing position technically and are arguing for alternative UX solutions. I’m not pretending it has to be perfect. Like I said, capabilities are great for creating a secure OS and writing more secure software more generally. But the threat model it’s protecting against is not software that phones home but against the size of the exploit opened up from a compromise.

Think about it this way, Android apps and iOS apps are largely sandboxed through a primitive capabilities system already, not super fine-grained capabilities but still the same concept. Would you care to claim that privacy and malware isn’t a problem on these systems or that the permissions model has meaningfully curtailed anything but the most egregious of problems?

Your editor doesn’t do it because handling, delegating, and slicing up capabilities isn’t a core part of the OS.

Firstly, VSCode runs on 3 major OSes that don’t have this capability and such software results the way it is partially because of targeting the lowest common denominator. Only a Fuscia first editor would do this.

Secondly, the editor also does it this way this because of reasons other than support within the OS because even with components it would need to design a capabilities model for extensions and a sandbox process to maintain the permissions - it’s much easier to just do the extensions in-process and not think about it.

> Only a Fuscia first editor would do this.

The difference is in that every single one of those other operating systems, applications just have network access. By default. No capability needed. This would not be the case in an OS centered around capabilities.

Multi-platform software develops integration with local OS APIs all the time.

Like I said, your thinking is way too black and white. Your inability to see a different world doesn’t make one impossible to exist. What is even the point of thinking this way? Your entire mindset boils down to “nothing can ever be better”.

We’re getting way of topic. What I said is that capabilities in an extremely limited to ineffective tool for the end user to try to enforce their own privacy and security because it doesn’t offer controlling for the semantics of what happens. For example, it would be very hard to take a fuschia system that simultaneously allowed for the Google assistant feature and not for telemetry / Google storing mic data that was sent to them.

I’d ask you refrain from personal attacks. That isn’t a fair characterization of what I said. All I said is that capabilities fundamentally doesn’t solve the phone home problem in many real world cases. I also highlighted that there are very real economic forces that must be accounted for in terms of understanding why software is architected the way it is. And no, there’s very few applications that have a completely different architecture per major platform which is what we’re talking about with capabilities. That’s very different from abstracting some platform-specific APIs here and there.