4 ms·
> “We understand that there is this persistent risk out there. We take this very seriously.” said every breach victim ever, after the breach.
by wglb 11y ago
> “We understand that there is this persistent risk out there. We take this very seriously.” said every breach victim ever, after the breach.
- phodo 11y agoTouche'. And this is followed by .... "The agency now plans to install two-step authentication across its network, Ms. Seymour said" Although TFA is not always the solution, there is no reason that every org should not institute this.
- fweespeech 11y agoYes, and we all know its bullshit. Especially if they were warned 4+ months in advance.
- stcredzero 11y agoHowever, something is clearly going wrong in terms of: 1) How security infrastructure is constructed in the first place 2) How security experts have communicated to and educated the public The way computers are put together and software is distributed makes them insecure by design. This puts the public (non-experts) in a position where they feel helpless to do the right thing. On top of this, a pervasive attitude amongst security experts is that laypeople are hopeless idiots. The science, medical, and public health fields have to deal with misconceptions and misinformation all of the time. When I compare outreach from those fields and computer security, I wonder if more shouldn't be done.
- superuser2 11y agoThe problem largely comes down to write secure software, a large part of which is stop doing manual memory management, to which the community responds "over my dead body." The problem isn't how security infrastructure is constructed, it's that people would rather buy "security infrastructure" and bolt it on to bad code. Laypeople have demonstrated themselves to be hopeless idiots with regard to 1) passwords and 2) infecting Windows installations for quite a while. Taking away the ability to accidentally screw these things up (2FA, sandboxing, walled garden app stores, etc) is a huge step forward.
- stcredzero 11y agoLaypeople have demonstrated themselves to be hopeless idiots with regard to 1) passwords and 2) infecting Windows installations for quite a while. The same thing can be said of certain plumbing and car repair tasks. In cases like this, industry and society should take the problematic aspects out of the hands of laypeople and put those things in the hands of tools/experts. Even highly trained groups like pilots have significant automation to aid them, where the population seems to suffer from "idiocy." (Stick shakers.) Such thinking and UX design is part and parcel of writing secure software.
- pdkl95 11y agoWhile I totally agree that memory management is a huge source of problems (learn Rust is next on my project queue), I suspect that the problem of "writing secure software" is not actually technical in nature. The best tools and methods in the world are only useful if they are actually used in practice. A better question is why people accept insecure software and then let the software industry blame the victims (like calling them "hopeless idiots"), when it was - as you say - the insecure software that was the actual problem. Dan Geer was right - you fix this with product liability. The tools (such as new languages like rust, etc) will happen automatically if you make the person who sold you software liable for the damage it causes. (standard "when used normally" restrictions apply, of course).
- wglb 11y ago> The problem largely comes down to write secure software, a large part of which is stop doing manual memory management, This is, unfortunately, not true. Writing in .net or in java or in php does not involve manual memory management, yet such web apps still contain game-over vulnerabilities. (I am looking at you, wordpress).