Schneier on Security: Manipulating Breathalyzers | Hacker News Reader

Schneier on Security: Manipulating Breathalyzers

schneier.com

19 pointsmblakele13 comments

Is this a joke?

A link to a 1 sentence post that goes to a bing search that somehow embeds a youtube video?

Here is the direct link: http://www.youtube.com/watch?v=eqlq25PRbko

Seriously, the linkjacking is one thing, but putting "Schneier on Security" in the title... what this says is, "this content is only interesting because Schneier has filtered it for us".

I linked to a couple of Bruce's stories a week or so ago, including one that was just a link, because a lot of the comments on his blog are well worth reading, often even better than here. The benefit of HN is the breadth of subjects, but if I had to choose one or the other, I would drop HN before schneier.com. Fortunately, I don't have to.

Not sure why is that exciting. There's a simple way to not get your result manipulated.

It's... shock - not drinking before you drive!

Interesting. Could you provide step by step directions on how to do this? I'm not really sure where to begin.

Sure - when you're about to drive, step back and undrink the drink.

Undrinking happens automatically after a dozen or so more drinks.

You're a Prolog programmer, I infer...

And then what? You tell the officer that you really haven't been drinking, and that they can trust you?

From the title, I'd thought somebody discovered a specific BAC & breath speed combination that'd crash the analyzer, or something. That would've been much more exciting.

Are you saying that no sober person has ever been tested? If someone is corrupt enough to manipulate a test to make someone test more inebriated, then they are corrupted enough to make a sober person test drunk.

No - I'm saying that this trick works, because the exit for air/alcohol is blocked. If you don't exhale alcohol to begin with, then this trick should not make any difference.

So if you're completely sober, then you may get "twice" the normal result - but it's still ~0.0

Your ignoring the real point here.

Reply on news.ycombinator.com