Text 16310, 76 rader
Skriven 2008-05-28 11:19:00 av Michiel van der Vlist (2:280/5555)
  Kommentar till text 16282 av Robert Bashe (2:2448/44)
Ärende: Analyses (was: Freedom)
===============================
Hello Robert,

On Tuesday May 27 2008 22:07, you wrote to me:

 BF>>> Then we can dsiscuss the reliability of those instruments.

 MV>> I think they are reliable enough for a quick first screening.

 RB> As someone who has actually performed air analyses for various
 RB> chemicals, I can tell you that a lot of people have _very_ exaggerated
 RB> ideas about the accuracy of such measurements,

Sure. I am not one of them.

 RB> particularly when only trace levels of a target substance are
 RB> involved. The results can vary several hundred percent depending on
 RB> the precise method and the exact circumstances of the analysis.

Yep. But with the breathalysers as used on the street, we are not aiming for an
exact measurement, we are aiming for a first screening,

And given that there is considerable margin between the limits as set by the
law and someone actually being a serious danger, I would say that an error of
100% is no problem for a first screening.

 RB> The police use such things not because they're accurate in any way,
 RB> but because they're simple to use even for the tecnically uneducated.

Of course.

 RB> The only problem is that the results don't mean much.

I think they mean enough to be usabale as a means of first screening. In my
very limited experience the things do not show any numbers at all. Just a
green, a yellow and a red light. A green light means no detectable traces of
alcohol. A yellow light is traces of alcohol, but not enough t warrant further
testing and a red light means you go through the mill.

They may have changed the equipment now that we have two limits. A 0,2 per
mille limit for drivers that have their license for less than five years and a
0.5 per mille limit for others.

 RB> Take a mouthful of gin, spit it out and wash your mouth with water.
 RB> Then take such a test. It will show you're highly intoxicated and
 RB> should be weaving all over the place.

Possibly. But that is not how he tests are taken.

 RB> And as for basing a decision on pulling you in for a blood test on an
 RB> air analysis for ethanol (or whatever reacts with the reagent
 RB> similarly), you can flip a coin and get the same results.

Even if that would be true, which I do not believe, it still serves as a
deterrent to drive under the influence.

What matters is that potential DUI's believe there is a reasonable chance of
getting caught.

 RB> The old-fashioned method of the policeman asking you to exhale in his
 RB> face wasn't much worse, and in fact was probably more sensitive

And gives rise to accusations of the policeman being influenced by personal
dislikes against the suspect. "I got pulled in because I am not white
caucasian.."

 RB>  - it also had the advantage that no questionable numbers were
 RB> generated, suggesting a degree of accuracy that simply doesn't exist.

The test may not be ideal, but can you come up with something better to keeo
drunk drivers off the road?


Cheers, Michiel

--- GoldED+/W32-MINGW 1.1.5-b20070503
 * Origin: http://www.vlist.org (2:280/5555)