In Wisconsin the Intoxilyzer 5000 is an approved method of testing pursuant to § 343.305(6)(b), Stats., and Wis. Adm. Code § Trans 311.04, and it is generally afforded a presumption of accuracy and reliability, see State v. Disch,, 119 Wis.2d 461, 475, 351 N.W.2d 492, 499 (1984); State v. Busch , 217 Wis.2d 429, 442-43, 576 N.W.2d 904, 909 (1998); In Minnesota, the intoxilyzer 5000 is presumed reliable if the proper procedure was used in running the test. State v. Dille, 258 N.W.2d 565, 567 (Minn. 1977) (citations omitted); see Tate v. Comm’r of Pub. Safety, 356 N.W.2d 766, 767-68 (Minn. App. 1984). This essentially shifts the burden of proof in a DWI/DUI case from the state to the driver to prove that the Intoxilyzer 5000 is in error. It is for that reason that many states having ongoing DWI/DUI litigation related to the source code of the Intoxilyzer 5000.

Before the source code, can be discussed, a brief and simplistic explanation of our “oracle” - the Intoxilyzer 5000 - is necessary. The Intoxilyzer 5000 is a machine with a complex set of functions. It is a computerized system that has a tube attached on one end for receiving breath samples, and a printer attached to the other end to record results. A driver breathes into the tube, the sample swirls around inside a chamber in the box, and the printer then displays the test results. Presumably what occurs is that the machine accepts a sample of air from a user and chambers that air. It then beams infrared light through the holding chambers where sensors on the other side of the chamber determine how much absorption of certain waves of infrared light has occurred. Based on that data, an analysis is made to determine the amount of alcohol contained in the sample to absorb that amount of light. Further mathematical calculations are done to determine the amount of alcohol in the underlying tissues in the lungs with respect to the determination of the amount of alcohol in the alveolar air.

The “source code, a computer program, is a coded set of instructions that control every mechanical function of the intoxilyzer 5000. That is true whether it is an instruction to turn the machine on, make the fan operate, or to make analytical analysis such as the rate at which alcohol absorbs infrared light and the mathematical formula comparing alcohol content in alveolar (“deep lung”) air to the blood alcohol content in the tissues below. As a result, in order to understand the inner workings of the Intoxilyzer 5000, it is necessary to have a copy of the source code and to have it independently tested for accuracy.

As a direct result, in many states, Minnesota included, defense attorneys have filed Motions seeking to compel disclosure of the source code for the Intoxilyzer 5000. The state does not have access to the source code and, when it is requested from the manufacturer, CMI, they are told it is a trade secret and that it will not be provided. In other words, nobody actually knows what goes on inside this machine except those who have viewed (or written) the source code. Yet, under law, everyone in the legal system must assume that the result is accurate and, to a great degree, indisputable. In other words, the response by CMI is the equivalent of saying – “trust us, it works.” Unfortunately, when “trust” results in the guilt or innocence of citizens, it is not so easily given.

That is particularly true when every computer owner knows, machines are subject to failures. The failure could be because of an incorrect code, a corrupted code or even subject to human error or human mischief. In short, machines are not perfect and the humans that program them are not perfect. What is particularly vexing is that the Intoxilyzer 5000 uses a microchip that is likely much less sophisticated than the one being used by your home computer. The testing process for a breath sample by the Intoxilyzer is run by a Z80 Microprocessor that utilizes an EPROM for program memory and handles all operator commands and interface devices. Simply put, this is not cutting edge technology. The Z90 microprocessor predates the original IMB PC. In fact, it was used back in the day for the Atari video game Pong. Your remember “Pong,” a game with all the complexity of two paddles and an electronic ball to bounce between the two. (See, www(dot)pchistory(dot)org)

The argument that a driver must have access to the source code in order to effectively challenge the results of a breath test appear to be gaining steam. In Minnesota, the state itself has sued that manufacturer of the Intoxilyzer 5000, CMI, Inc., arguing that based on its contract with CMI the state is the actual owner of the source code copyright. In suing CMI, the state of Minnesota avers that it owns the intangible copyright to this source code and demands, inter alia, specific performance of its contract and its right to possess (and copy) a tangible version of the intangible, intellectual property that the state contracted and paid for. The issue regarding the disclosure of the Source code as part of a DWI/DUI case in Minnesota is proceeding to the State Supreme Court in the case of State v. Underdahl and will be heard later this year in 2008. Ultimately, the result may provided those accused of DWI offenses in that state with some ammunition to challenge the results of the modern day “oracle” - the intoxilyzer 5000. Until the matter is decided, Motions seeking to compel disclosure of the source code or suppress the breath test result, remain an important and viable part of the DWI defense arsenal. Such motions should be considered in each case.