On Tue, Mar 10, 2015 at 8:23 PM, <travis+ml-lang...@subspacefield.org> wrote:
> Incidentally, has anyone noticed that the recommended path to security > sometimes seems to be "handcuffing" oneself, for some definition of > handcuffing and some definition of oneself? For example, standard > langsec recommendation, SELinux, sandboxing, MMUs, Trusted Computing > all involve restricting power. I see those things (except perhaps SELinux, which I have a hard time taking seriously) as providing much more freedom than they take away. Privilege separation enables mobile code and for multiple people to share 1 machine. MMUs enable virtual memory, which enables the programmer to pretend to have free reign over the entire address space and frees the programmer from having to worry about clobbering other programs or the kernel. Langsec allows us to have simple interfaces that are more obviously safe, reduces maintenance costs, and generally results in higher performance. We're all much happier with loops, function calls, and try/catch than with the more "free" goto. :) "The really important kind of freedom involves attention, and awareness, and discipline, and effort, and being able truly to care about other people and to sacrifice for them, over and over, in myriad petty little unsexy ways, every day." — David Foster Wallce _______________________________________________ langsec-discuss mailing list langsec-discuss@mail.langsec.org https://mail.langsec.org/cgi-bin/mailman/listinfo/langsec-discuss
