That doesn't resolve the issue of an application being compromised and an attacker gaining access to everything in that environment. Security against remote compromise and fine-grained containment certainly matters despite coarse-grained isolation chosen by the user higher up.
Conversation
For the secure messaging example, it can be isolated per contact, and handling things like audio / video decoding for video calls can be isolated, as can cryptography, etc. Finer grained isolation than a group of applications for a certain identity / task is very important.
1
I think this approach is really less about containment and more about reducing the complexity of data formats exchanged between subsystems. Even the strong isolation of Qubes relies on the idea of simplified formats used at critical points.
1
Isolating per-contact in a messaging client, per-site in a browser, etc. is applying the same principle of QubesOS at a fine-grained level using existing privacy/security boundaries. Since they're existing boundaries, it doesn't require the user to do anything or be aware of it.
2
OK re per-contact-site isolation.... Now, demote "kernel" to "services" and look to hypervisor+qrexec as the actual kernel.
1
There are implementations of fine-grained isolation within applications using different mechanisms than OS sandboxing. Architecture-level virtualization is one possible approach and has pros / cons, as do other approaches like a higher-level virtual machine, etc.
1
1
The isolation between sites in a browser or contacts in a messaging app are good examples of existing fine-grained trust boundaries to reinforce. There are a lot of other examples and reinforcing those can improve security for a billion users with no more work on their part.
2
IDK if that's literally true. The move to https was itself a form of site isolation & industry was tinkering with hiding URLs altogether when they did about-face out of necessity. We might not have Chrome internal isolation today if not for trend reinforcing site identity in UI.
3
Representation of https became weaker (location semantics "scare" users!) in browser UIs before they became stronger. Location bar became explicitly a part of security context, lock icons returned, plus EV certs and domain highlighting.
1
Lock icons and EV are on the way out. The default will be HTTPS, anything not HTTPS will have scary warnings and the only identity will be the domain since that's the only identity that ever worked / mattered.
It will probably change further down the road, but that's the roadmap in browsers like Chrome and Firefox for the near future.
You sound like me 12 years ago. 🙂 Seriously, https (and EV) are valuable for some threat models. And they can be subverted. My mention of them was not intended to promote them as "only identity that ever worked", etc. Its just an example of isolation in the UI.
1
HTTPS is obviously valuable. EV doesn't accomplish the intended purpose because company names aren't unique and positive security indicators are near universally ignored by users. If a user doesn't change their behavior based on it being missing, it accomplishes nothing.
1
Show replies