March 29, 2024

Breathalyzers and Open Source

Lawyers for 150 Floridians accused of drunk driving have asked a court to order the disclosure of the source code for software running in the breathalyzer machines used by police to analyze their blood alcohol level, according to a Tom Sanders story on vunet.

The defendants say they have the right to examine the machines that accused them, and that a meaningful examination requires access to the machines’ software. Prosecutors say the code is a trade secret.

The accused are right that one needs the code to understand fully how the machines work. The machines consist of sensors, a user interface, and control software. The software is the “brain” of the machine, and it is almost certainly involved in the calculations that derive a blood alcohol value from the sensor readings, as well as the display of the calculated value. If the accused have the right to fully examine the machines – and the article says that they do under Florida law – then they should see the source code.

Contrary to the article and some other commentators, this is not a dispute over whether the software should be open source. The accused aren’t seeking to open the software to everybody; they only want it opened to their legal teams.

There are standard practices for handling trade-secret information that must be turned over in court cases. A court will typically establish a protective order, which is a kind of nondisclosure agreement covering secret material that is turned over by one side to the other. The protective order will require parties to keep the information secret and to use it only for purposes related to the court proceedings. Typically the information can be turned over to a limited number of expert analysts who have also signed the protective order. Documents containing secret information are filed under seal, and testimony about secret matters may take place in a closed courtroom.

So this issue is not about open source, but about ensuring fairness for the accused. If they’re going to be accused based on what some machine says, then they ought to be allowed to challenge the accuracy of the machine. And they can’t do that unless they’re allowed to know how the machine works.

You might argue that the machine’s technical manuals convey enough information. Having read many manuals and examined the innards of many software systems, I’m skeptical of such claims. Often, knowing how the maker says a machine works is a poor substitute for knowing how it actually works. If a machine is flawed, it’s likely the maker will either (a) not know about the flaw or (b) be unwilling to admit it exists.

If the article’s description of Florida law is correct, this seems like a pretty easy decision for the court.

Comments

  1. in my opinion any software that is used to create evidence in a criminal case should be open to scrutiny. When people’s lives and freedoms are at stake manufactirers should be held to a high standard. recently there have been plenty of instances where breathalyzer software has been poorly designed and implemented, leading to unreliable test results. Despite the bad test results the outcomes are taken as gospel by both police and juries.

  2. I was recently stopped, after about a one and one-half hour drive, was sleepy and pulled over. I tested positive for alcohol on the Alcotest 7110 MK III. I had been in the sun all day using sunscreen which contains
    2-propanol. I understand that 2-propanol has a significant effect on the test results of the Alcotes 7110. Any additional information available ?

  3. As a former owner of an authorized service center for intoxilyzers I would like to point out a couple of facts. It is not possible to blow anything chewing excel gum, gummi bears, or anything else for that matter and get any kind of substantial reading.

  4. Can anyone provide more info on similar challenges in New Jersey? Rulings, pleadings, news stories, whatever… ERIC

  5. The Florida issue seems to be based around the fact that the Florida regulating authority did not adequately test the changes that the manufacturer made to the instrument -subsequent to the intitial instrument approval. If the state had adequately tested the instrument, they and the manufacturer would not be in the current situation.

    To the defence of the manufacturer, I struggle to believe that any programmer can read the source code from an instrument such as a breathalyzer and know how it will affect the performance of the instrumentation. Testing the instrument itself is the only scientific method to determine if the code is good or bad. An experienced programmer can only offer conjecture about how the instrument will perform. Physical testing is the only real evidence to prove or disprove the instruments capability and this can be done without the aquisition of the source code.

    Has the State or the manufacturer declined to offer the accused access to the instrument???

  6. […] You might argue that the machine’s technical manuals convey enough information. Having read many manuals and examine the innards of many software systems, I’m skeptical of such claims. Often, knowing how the maker says a machine works is a poor substitute for knowing how it actually works. If a machine is flawed, it’s likely the maker will either (a) not know about the flaw or (b) be unwilling to admit it exists. Link […]

  7. […] Story here. [Jack Chin] […]

  8. This is part of the overall problem with prosecution today, whether in drunk driving or other areas. Prosecutors, judges (mostly former prosecutors) and politicians aren’t satisified with conviction rates in the high 90s. They keep tilting the field against criminal defendants. And the public is blissfully unaware. Until they get pulled over after three drinks, and are arrested for drunk driving.

  9. Tom Sanders told: Open Source means “third parties are granted unlimited access to the source code”

    That is not so, in no way.
    GPL puts limitations on what You can do with source code, BSD, MPL, etc – all them puts limitations. Only the code that has no license attached – has no limitations.

    > The defending lawyers in this case are asking for that
    I doubt! Do the ask for the right to change the code ? To sell the code ? To publish it? To show it to any man except for one settled by judge ?

  10. update II:

    I spoke to the attorney in FL today handling the case with the successful result, and he said that there is a case on the other side of the state with the opposite result. The State is likely appealing his case, as will the defendant in the unsuccessful case.

  11. Welcome to New Jersey! We have this fight going on right now with the machine that is being implemented throughout NJ — the Draeger Alcotest 7110 MKIII-C. See the machine at my website, http://www.dwi-nj.com/parts/whichmachine.html, and a general explanation of the machine at http://www.dwi-nj.com/parts/breathtest.html. We are challenging this machine in Middlesex county and several other counties to obtain the source code for this machine.

  12. update:
    motion granted… cases likely to be thrown out.
    http://www.vnunet.com/vnunet/news/2145410/judge-gets-bottom-possible

  13. drunken computer nerd says

    mature seems to think responsibility/accountability should only apply to the accused. Aren’t police officers and the corporations that serve them equally capable of making mistakes ? Should they not have to own up to them?

    IMHO the real problem lies in the number of people who admit guilt to a crime they aren’t guilty of, through ignorance of their rights, lack of resources, or faulty evidence that is deemed unassailable.

    The courts blind trust of these corporations (who have an obvious conflict of interest in this case) borders on judicial fascism.

  14. I have to disagree with your “not about open source” assertion. Open source doesn’t have to mean: make available under the GPL, it simply means that third parties are granted unlimited access to the source code. The defending lawyers in this case are asking for that. Whether their request will be granted is another point.

    You could rightfully argue that a study by a trusted third party can solve this dispute. However, the Florida statue is very specific in the rights of accused drunk drivers, granting them direct access to the technical manuals and such – not through a trusted third party.

    Btw… the name of the website is vnunet, not vunet (confusing name… I know… blame it on the VNU parent company)

  15. Johnson Moore says

    “most cops are too lazy to take you to the hospital for the blood sample..if you sit in his car for 5 minutes before you blow..its grounds for dismissle. and hopefully you witness him putting on a fresh blow pipe with out touching the blow pipe with his dirty hands..health safety regulations..”

    If they put you in a car, before formally charging you and the officer closes the door YOU are under arrest. When you get to court the entire case will be dismissed automatically by the Judge when these facts are brought to his attention. They cannot detain you without telling you, you are being “detained” or arrested. I know from personal experience. Also, you are entitled to a “witness” for your test if taken downtown to blow. If your “witness” is to be your attorney then they must wait, or charge you on the merits of the original stop/detainment. I know, beat that one too.
    They cannot take your blood without your permission, I know a friend who was in a car accident and the police took blood while they were unconscious. Thrown out in court as well. Fact is, if you know the law and are willing to stand up for YOUR rights nothing sticks…We pay for the “service” of being arrested anyway in state taxes!

    ASK O.J.

  16. I think it’s obvious that finger print matching, dna matching, breathalyzer testing and any similar procedures that can result in arrests and convictions should be judged against open standards and procedures. Perhaps the breathalyzer market has developed nicely by having trade seret protection, but that’s wrong in the long run.

    Would you like some lab to tell you that your DNA has been matched to a rape, but they will not tell you how the match was calculated?

  17. I spent a career as a biochemist. I would not want to depend upon a quantitative value produced by any field machine that has not been calibrated and tested known controls. This should be done within minutes of the actual test of the unknown sample. Especially when tested by a lay person (which would include a police officers).

    In many states a follow up blood test is required. A video of the sobriety tests is also common. Absent the examinee’s other physical impairment or disability, the video is probably the most conclusive evidence possible.

  18. To be clear:

    I am not advocating letting guilty drunk drivers go unpunished. If the machines are reliable, they should be able to withstand scrutiny. If they are not reliable, then the police should stop using them and switch to a better brand of machine, or to blood tests, or to other forms of observation.

    I don’t think it’s fair to surprise a police officer, on the witness stand, by asking him to interpret the source code. But I do think it’s reasonable to give a defendant the right to have an expert examine the machine’s source code before trial.

    I’m not convinced by once-a-month pro forma tests. I’m certainly not convinced by a power-on diagnostic. These will detect a machine that needs repair, but I wouldn’t trust them to detect a design flaw. It’s well known that this sort of testing isn’t adequate for software products.

    Has the county or state ever had a chance to examine the machine’s source code, or do even they not have the right to look at it?

    Again, if the machine is actually reliable, none of this will matter. Some defendant will spend a fortune having the machine’s software analyzed, and will be convicted anyway.

  19. I first want to address this quote:

    “”Did you know that these breath alcohol analyzers will blow over the limit if you chew a piece of Excel gum. Apparently, this is the little dirty secret they don’t want you to know about. If anyone has access to one of these devices they should test this out by chewing a piece of this gun and blowing into an analyzer.

    If this is true this means these devices cannot be use to testify any more than lie detectors can be used to gather evidence that it admissible in court.””

    I seriously doubt the validity of that statement.

    As for wanting to see the source code. That is a hoop that a police officer on stand should not be subjected to jump through. The intoxilyzer is already tested for accuracy each month by an agency inspector according to strict state regulated standards and these results are documented & then reported to and kept on file by the Florida Department of Law Enforcement.

    Also, when it is turned on the intoxilyzer runs a diagnostic check and an air blank.

    Police officers also must become certified to be able to use it. They must successfully pass a written test, demonstrate knowledge of it, be able to explain the process, and show proficiency.

    I know several people who have been victims of a drunk driver, two of them are is no longer here today… I appreciate the efforts of law enforcement to make the streets my family travels on a little bit safer and would hate the focus of the trial to turn from the impairment of the defendant to the source codes of the intoxilyzer. Every year there seems to be less and less accountability for peoples actions.

    Making it hard for police to keep the streets safe will only discourage them from doing just that. The result of that will be an increase in DUI related traffic fatalities because guilty people walk because they look for silly loopholes. Sad…

  20. Jack Brown says

    “Frankly, I think the US should go to the way Japan does it… ANY alcohol in your system and you’re hosed. Makes it really easy and I guarantee it will reduce the problem to near zero.”

    And, like USELESS arrests for pot-users, this would fill our jails with otherwise law-abiding citizens. Zero Tolerance in any form is a simplistic “fix” that results in many unintended consequences… like taking jail beds from serious criminals to make way for light drinkers.

    Not all that wise of a solution to a societal problem that has DEMONSTRABLY improved over the last 2 decades. The totalitarian approach is not the answer.

  21. What ever happend to personal responsibility/accountability? They can’t accept the fact that they were driving drunk and own up to their mistakes. No… they have to burden many other people and make it even harder to get the drunks off the road. Frankly, I think the US should go to the way Japan does it… ANY alcohol in your system and you’re hosed. Makes it really easy and I guarantee it will reduce the problem to near zero. No maybes, no court battles… you either drank or you didn’t and that’s easy for a machine to pick up accurately. None of this crap surrounding a .081% and a “margin of error” defense. Positive or negative… problem solved. Bet you’ll think twice about that mouthwash now, eh? Alcohol is alcohol… and you know what? Ask anyone in Japan if they have a problem with drunk drivers? And it’s a country where they have beer vending machines on every street corner… no, I’m not kidding.

    Instead, we Americans seem to think we have a right to get sloshed and then contemplate with clouded judgement whether we might be sober enough to drive and/or pass a breathalyzer test. And when we get nailed, any lame excuse is fair game as long as some idiot judge or jury buys off on it.

    Plain and simple… if you take a single sip of alcohol (or any other chemical substance for that matter), the onus is on YOU to deal with the consequences. Grow up, people. Pretty much anyone that pleads not guilty to a crime they know they are guilty of is a pathetic excuse for a human being.

  22. ****If you are ever stopped and you might be only slightly over the limit, be sure to ask for a blood level from the vein.****

    most cops are too lazy to take you to the hospital for the blood sample..if you sit in his car for 5 minutes before you blow..its grounds for dismissle. and hopefully you witness him putting on a fresh blow pipe with out touching the blow pipe with his dirty hands..health safety regulations..

    but the problem with these machines are always temperature causing the sensors to deviate from the 70 degree balance.

  23. Dan you are assuming guilt in your response, they are fighting to see the code because they believe they are not guilty.

    http://www.duiblog.com/2004/10/17#a6

    lots of good information about DUIs at that sight and why you should care.

  24. Umm.. if this goes thru, they better allow the source code for the voting machine’s to be examined as well.

  25. one of my friends used to sell those devices to lots of people including bars and police community. I got to play with one for an afternoon. try a few spritzes of binaca if you truly want to see something hilarius. I blew a sober 6.00 whee!!! also eating (recently) many low carb sweets will trigger them I was able to blow 2.25 with just having swallowed a mouthfull of sugar free gummi bears. Oh and my favorite, my judge-sober mother (teetotaller never even had a drinnk in her life) blew a 1.20 after having consumed a hotdog with mustard and a piece of cantelope. couldn’t figure which item triggered the device but they are not effective. If you are ever stopped and you might be only slightly over the limit, be sure to ask for a blood level from the vein. Ask the officer are these accurate 100% of the time on 100% of the people. I f they say yes then blow away, the camera caught them giving you a free out, if not then go to the hospital, have them call your family doctor to come take your blood. it buys the only thing that can lower your bac… time.

  26. so what do we have a bunch of drunken computer nerds driving drunk in this state or something?

  27. Jiri, I’m all in favor of accountability and oversight. But there are innumerable ways of providing these things without letting drunk drivers off scot-free. Audits, spot-checks, complaint investigations–a whole gamut of methods can be put to use to detect and eliminate problems with the system. In fact, I can’t think of a worse approach than giving a pass to anyone whose lawyer can make a sufficiently big deal, rhetorically speaking, out of some alleged niggling doubt about the evidence. Remember the O.J. Simpson trial?

  28. Did you know that these breath alcohol analyzers will blow over the limit if you chew a piece of Excel gum. Apparently, this is the little dirty secret they don’t want you to know about. If anyone has access to one of these devices they should test this out by chewing a piece of this gun and blowing into an analyzer.

    If this is true this means these devices cannot be use to testify any more than lie detectors can be used to gather evidence that it admissible in court.

    What is sauce for the goose is sauce for the gander.

  29. Dan,

    the point of “right to examine” is not so much fairness to the individual defendant as a general deterrent from the police getting rigged equipment.

    Beating a speeding ticket by paying a lawyer is neither useful nor dangerous – the lawyer will almost certainly cost more than the fine anyway (in money, stress, time…).

    However, *somebody* needs to have the right to inspect the equipment the police is using. You might argue whether defendants are the best people for the job, but it is a job that needs doing. If you want to argue against defendants doing it, it’d probably be a good idea to suggest someone else to have the right…

    Jiri

  30. So this issue is not about open source, but about ensuring fairness for the accused.

    Actually, it’s about neither. The “right to examine” ruling isn’t about “fairness”–it’s a loophole allowing anybody with money and savvy to beat the system.

    In Washington state, there have long been elaborate rules on the books requiring the police to jump through a bunch of onerous hoops to make a speeding ticket stick. Like the “right to examine” rule, they’re ostensibly protections for falsely accused speeders. In practice, though, their effect is to “protect” anyone with the money and savvy to hire the right lawyer. Those fortunate few can beat any ticket, whether they were speeding or not, while everyone else gets zero benefit from the “protections”.

    (The exception is one county in Eastern Washington which has decided to treat speeding tickets as a profit center, by investing substantial money and manpower in ensuring that its speeding tickets are legally unassailable–and then making its expenses back, with considerable profit, by enforcing the speeding laws ruthlessly.)

    Of course, a law that allows anyone to beat a speeding ticket by paying a lawyer probably isn’t a society-wide catastrophe. But one that allows people to beat DUI convictions by paying a lawyer will almost certainly cost a significant number of lives. And in practice, the “right to examine” law–not to mention its open-ended expansion through litigation, along the lines of this latest “open source” argument–will do just that.

  31. There already are well-established procedures on how to handle this kind of thing.

    The basic idea is that IF the judge agrees with the defendants, they are allowed to examine the secret materials, but are on the hook for keeping them secret. Discussions of the secret materials during the courtroom proceedings are sealed, ie. the public-record courtroom transcripts are edited to remove the secrets.

    In other words, the source does not become “open” in any sense of the word.

    NB the word “IF” — the defendants are going to have to make an awfully good case that all the previous challenges to the accuracy of machines like this were not good enough, and that in order to really ascertain the machines’ fairness and accuracy a black-box test is not good enough.

    (Interestingly, some of them came into play during some of scientology’s many many lawsuits. Here’s one web link that seems particularly appropriate: )

  32. If I were the defendant and I didn’t want to show the code, I’d propose to let an independent lab run tests on the machine.

    After all, the code could be perfect and the machine could still fail: there could be electrical, chemical or mechanical failures not related to the software.

  33. It would be nice if black-box testing were sufficient to exercise software-controlled devices to determine whether they worked reliably. Then we could all go happily about our lives using all the bug-free code that would be the only kind released by responsible vendors.

  34. Michael Langford says

    > Black-bos testing is usually enough for such
    > an analysis – one need not see the code to
    > assert whether the device is reasonably accurate or not.

    As a researcher on device-simulation, I’ll say you’re off the mark.

    Black box testing is only good enough if you know the exact cirumstances the device was used in. I would say that there are too many variables for that scenario to be reliably demonstrated, especially since you’re talking about chemical sensors.

  35. It really depends on the interpretation of “right to fully examine” the device. Black-bos testing is usually enough for such an analysis – one need not see the code to assert whether the device is reasonably accurate or not. Notwithstanding, just saying “read the tech. manual” is just not enough.