| likes | books, literature, maths, videogames, physics, mechanisms, maps, calculators, computers, water, sherbats, coffee, teas, cats, fish, archives and libraries |
| dislikes | escalators and dishonesty |
| likes | books, literature, maths, videogames, physics, mechanisms, maps, calculators, computers, water, sherbats, coffee, teas, cats, fish, archives and libraries |
| dislikes | escalators and dishonesty |
True
(3 weeks left for me 🎉🎉🙌)
there are really two major schools of thought in the security world.
there is the school of thought that pursues secure-by-design approaches. concepts like object capabilities come out of this way of thinking.
things which cannot be built in a secure-by-design way are hardened with mitigations to bring the thing as close as possible to being secure-by-design. concepts like PaX, pledge, landlock and seccomp come out of this way of thinking: if there is an attempt at aberrant behavior, deny it and let the program crash.
there is another school of thought that pursues interventions as the primary line of defense. concepts like antivirus, runtime anomaly detection, and so on come from this way of thinking.
the problem is that interventions can be bypassed: eBPF can be turned off, and trying to recover from an incident after it has already occured winds up being more intrusive than simply not allowing the undesired behavior to begin with.
interventions are easier, though, which is why most of the industry is focused around selling interventions.