3 ms·
Why stop at the browser? Why not do this for every app? And sure, this makes exploiting the desktop from the web more difficult, but only by requiring an addit
by chatmasta 10y ago
Why stop at the browser? Why not do this for every app?
And sure, this makes exploiting the desktop from the web more difficult, but only by requiring an additional exploit in the chain. I'm sure there will be vulnerabilities in the virtualization that hackers will pair with edge vulnerabilities.
This is a good step so long as they continue to patch browser vulnerabilities at the same pace. But if they use it as a crutch ("oh that vulnerability is no big deal because edge runs in a VM!") then it quickly becomes a problem.
- Analemma_ 10y agoMost applications don't have anywhere near the huge Web-facing attack surface that the browser has, and have to do a lot more communication with the host OS than the browser does, so the tradeoff is less worth it. However, I think something like this will be slowly making its way to more applications as time goes on. The browser is just the most obvious first test subject. I'm not really worried about it being used as a crutch. ASLR and stack canaries didn't stop people from fixing application bugs, because people know that no mitigation technique is perfect, and they're supposed to be a last resort if the application fails.
- drinchev 10y agoThis sounds awfully similar to iOS sandbox mechanism.
- compsciphd 10y agosomething like this? https://www.usenix.org/conference/usenix-atc-10/presentation/apiary-easy-use-desktop-application-fault-containment-commodit https://www.usenix.org/conference/usenix-atc-10/presentation...