On Sat, 10 Mar 2012 16:08:44 +0000 lkcl luke wrote: > if such a strategy (execution of assembly code) is OUTRIGHT BANNED and > will NEVER BE CONSIDERED, then and only then can the security model be > dramatically simplified.
I think this notion also applies as a whole and the key is flexibility for the user. Android probably wouldn't have been such a success if it wasn't for the huge inherited java legacy and ease of closed source app creation and installation. This same success holds it back to some degree for serious work. This shouldn't be prevented but giving the flexibility to users such as enterprise to prevent that and allow users to install verified open source apps from trusted sources would help enterprise keep they're users happyish and allow phones to be used more seriously safely, once the mobile networks sort they're encryption etc., out of course. p.s. Whilst I agree with all of these efforts and allowing all code for success. It should be remembered when designing any permission system that you are only raising the bar as high as you can and these disussions have been encouraging but the traditional method of choosing what local code exists is the most safe method but could also benefit from a permissions system. Allowing arbitrary (closed source) and functional local programmable code execution has never and probably never will be successfully isolated, especially with a bloated Linux kernel. _______________________________________________ dev-security mailing list [email protected] https://lists.mozilla.org/listinfo/dev-security
