home

DUI Laywers Seek Intoxilyzer Source Code

by TChris

The company that makes the Intoxylizer -- a breath testing device that's popular with law enforcement agencies -- refuses to release the source code to lawyers who want to verify the machine's accuracy. The company's unwillingness to give lawyers a behind-the-scenes look at the machine's workings may imperil DUI prosecutions that depend on a breath test.

The company that makes the Intoxilyzer refused to reveal the computer source code for its machine because it was a trade secret. A county judge [in Florida] tossed out [an] alcohol breath test - a crucial piece of evidence in a DUI case - and the ruling was upheld by an appeals court in 2004.

Since then, DUI suspects in Florida, New York, Nebraska and elsewhere have mounted similar challenges. Many have won or have had their DUI charges reduced to lesser offenses. The strategy could affect thousands of the roughly 1.5 million DUI arrests made each year in the United States, defense lawyers say.

Florida law gives defendants the right to request "full information" about the testing device. The strategy seems to be working in states that provide similar discovery rights.

Breath testing devices are based on dubious assumptions, making it all the more important to have complete information about the test's validity.

"It seems to us that one should not have privileges and freedom jeopardized by the results of a mystical machine that is immune from discovery," Florida's 5th District Court of Appeal ruled in Muldowny's case, which resulted in his charges being reduced to reckless driving.

Some Florida lawmakers have more interest in protecting the manufacturer and in preserving the myth that breath tests are infallible than they have in assuring that convictions are based on reliable evidence. They've introduced a bill to exempt source codes from the disclosure requirement.

< Judge Issues Secret Ruling in Case Challenging NSA Wiretaps | A Veteran Writes Bush, Returns Navy Wings and Bars >
  • The Online Magazine with Liberal coverage of crime-related political and injustice news

  • Contribute To TalkLeft


  • Display: Sort:
    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#1)
    by Talkleft Visitor on Sat Mar 11, 2006 at 12:26:50 PM EST
    While I develop open source software, and I am a fervent believer in open source software, and I strongly believe that government should be using open source software whenever and wherever it can, I don't think that examining the source code here is necessary nor sufficient. I do think it would be very important to examine the source code, BUT, I would be more interested in their testing protocols and all of their test results, I think that could tell just as much about what the device can and cannot do. The law treats science in a fundamentally broken manner. Judges with no technical expertise arrogantly (ignorantly?) write all sorts of stupid decisions that make all sorts of scientific judgements. A former employee of International Airport Centers, who is currently embroiled in a legal dispute with them, returned his company laptop as required. Hoping to find incriminating evidence, I.A.C. attempted to retrieve deleted information from the laptop in question with no success. This employee had beaten them to the punch. He had used 'secure delete' software, in order to make sure nothing could be recovered. He is now being charged with a violation of the Computer Fraud and Abuse Act. In this case, Judge Posner wrote that Ordinarily, pressing the "delete" key on a computer (or using a mouse click to delete) does not affect the data sought to be deleted; it merely removes the index entry and pointers to the data file so that the file appears no longer to be there, and the space allocated to that file is made available for future write commands. Such "deleted" files are easily recoverable. But his statement is just not true. For years and years and years, deleting a file meant the file was deleted. On other operating systems, or in secure environments, deleting a file meant the file was g.o.n.e. And even on Windows, Windows can be set up in a normal everyday fashion so that is true. I know, I work in a secure facility that requires such deletion efforts. Posner's flat out statement is not true, but by assuming it is, he makes the argument that the employee's placing a secure deletion program on the machine is an act of malice. It's not an act of malice. That and the use of encrypted file systems should probably be everyone's day to day operation of their computer and how they protect their privacy, and their client's privacy, and their employer's privacy from theft. And in fact, I am required by my employer to operate my computer in exactly that manner. Near as I can tell, law is easy, but real world experience is hard. I think that every higher court should have a majority of lawyers on it, and a significant minority of other accomplished individuals: physicists, biologists, mathematicians, restauranteers, chemists, engineeers, actors, etc., on a 2 year term.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#2)
    by Talkleft Visitor on Sat Mar 11, 2006 at 01:41:16 PM EST
    I habitually use the secure delete option on my Macintosh. Then I sometimes worry that the gov't will think I'm hiding child pornography or terrorist communications.
    I would be more interested in their testing protocols and all of their test results, I think that could tell just as much about what the device can and cannot do.
    If their testing was done properly, you're right. But a lot of companies don't test their products adequately (that's what customers are for!). I wonder if a company could claim that their testcases and results were trade secrets?
    I think that every higher court should have a majority of lawyers on it, and a significant minority of other accomplished individuals:
    But how would these accomplished individuals be determined? Judge Posner probably got his information from an "expert". There are those who claim to be experts and those who really are experts. As far as I can tell, courts don't distinguish between the two. Ultimately, in my opinion, if a device is going to be used by the gov't to regulate peoples' behavior, the source code (as well as the specification and testing) should be available for public review. Imprisoning people or measuring peoples votes based on the magic sayings of secret machines is less than democratic. Companies that don't know how to make profits with that model should stay out of that field.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#3)
    by Lora on Sat Mar 11, 2006 at 03:40:01 PM EST
    Too bad we can't throw out the results of elections when e-vote companies refuse to reveal their source codes, don't subject their hardware and software to adequate testing, etc. (sorry, couldn't resist.)

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#4)
    by Johnny on Sat Mar 11, 2006 at 03:42:33 PM EST
    The stupid breathalyzer exists to help convict people of drunken driving after they have demonstrated the physical ability or in-ability to operate a motor vehicle via "roadside tests". It goes hand in hand with the even stupider, purely arbitrary, BAC limits set forth. Remember you were not drunk until 0.10? Now you are drunk at 0.08! Holy cow! Pass all physical testing and they can still claim you are incapable of driving based purely on a chemical measurement! Insanity!

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#5)
    by Talkleft Visitor on Sat Mar 11, 2006 at 04:58:35 PM EST
    But a lot of companies don't test their products adequately... That's why I want to see their testing protocols. But how would these accomplished individuals be determined? Probably from their peers. Let the National Academies of Scientists select candidates in much the same manner that they provide candidates to congressman as advisors. They provide this for medicine, science, and engineering. That would make me happy. I'd say anyone like that is just as qualified to be on an appeals court as any judge that makes it through the political battles.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#6)
    by Sailor on Sat Mar 11, 2006 at 06:41:01 PM EST
    JPF makes a point (no pun intended;-). We rely on judges to judge the law, shouldn't we rely on scientists, the scientific method, and peer review to judge the science? Since this has never happened with fingerprints, I'm guessing the answer is 'no.'

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#7)
    by Talkleft Visitor on Sun Mar 12, 2006 at 01:15:17 AM EST
    For the rational study of the law the black letter man may be the man of the present, but the man of the future is the man of statistics and the master of economics. - Oliver Wendell Holmes, The Path of the Law (1897)
    Not sure what it means, but it seems somehow relevant.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#8)
    by Talkleft Visitor on Sun Mar 12, 2006 at 09:37:43 AM EST
    Whether examination of the source code is necessary or not is irrelevant at this point. The source code must be disclosed so that the attorney(s) can decide what if any further examination is necessary. Basically, due process mandates the disclosure so that people can confront the witness(s) against them. Here, the machine is the witness. So, people should be entitled to cross examine that witness and the code may or may not be a fundamental part of that cross. Sometimes these defenses come down to a very narrow margin of error. A few percent can mean the difference between guilty or not.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#9)
    by Talkleft Visitor on Sun Mar 12, 2006 at 09:40:13 AM EST
    Actually, I don't just want the source code. I want to know who wrote it. They are/should be subject to cross as well.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#10)
    by kdog on Mon Mar 13, 2006 at 06:56:29 AM EST
    Besides just the code, how often are the machines calibrated and maintained? When I served on a grand jury a couple years back, I heard lots of cases where people blew 0.01 or less over. I asked the testifying officers in these cases how often the machines are checked for accuracy, and if there is a margin of error. They didn't know, saying that was the responsibility of others within the department. If these machines are being used to send people to jail, the public must have faith in their accuracy. I, for one, would like some answers to these questions if these machines will be used to decide a man's freedom.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#11)
    by Johnny on Mon Mar 13, 2006 at 07:09:00 AM EST
    I, for one, would like some answers to these questions if these machines will be used to decide a man's freedom.
    That's exactly why I feel the field sobriety tests should be the final determiner. And that's exactly why the state has decided that some idiotic, purely arbitrary chemical count should be the determiner. I don't know how these things work, but if someone blows a .01 over the "limit", that is greater than a +10% difference from a theoretical nominal. I would like to think the manufacturer holds tighter standards than that. Unfortunately, this is probably untrue. Just outlaw the stupid things. If a cop can't prove a person is drunk by making them perform tasks supposedly designed to demonstrate ability behind the wheel, then they need to let them go or someone needs to design different tests.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#12)
    by kdog on Mon Mar 13, 2006 at 07:31:43 AM EST
    One problem with that Johnny...some cops lie. Say the wrong thing at a traffic stop and all of a sudden you are "drunk". I think the best we can hope for is a combination of the two. Make the officer determine a suspect is drunk through field tests, then go to the breath test to confirm the officers findings. IMO, one without the other should be insufficient to convict.

    Re: DUI Laywers Seek Intoxilyzer Source Code (none / 0) (#13)
    by Talkleft Visitor on Sat Apr 22, 2006 at 09:59:29 PM EST
    Without a reliable and independantly verifiable method of testing a persons alcohol content close to the time of driving, IMHO it should always be insufficient to convict a person based solely on an officers observations, whether casual observations or application of NHTSA standardized field sobriety tests. The problem with field testing is that even under clinical conditions such testing is, as its very best, around 90 percent accurate. I have seen officers self reported horizontal gaze nystagmus logs reporting accuracy rates as high as 98 percent, which defies logic and credibility... but without an accurate BAC test it is impossible to refute the officers observations, only to create reasonable doubt. Without the source code, which CMI refuses to divulge, it is impossible to know if the test is actually accurate in and of itself.