More on the New Intoxilyzer 9000

     The COBRA system (Complete Online BReath Archives) is a CMI manufactured data storage and management program for the Intoxilyzers. Every test or attempt at operation is, in theory, stored within the individual machine’s computer memory and retrieved to be stored and used by the COBRA (or other database) system.  When complete and intact, the database can be an invaluable tool for evaluating the overall performance of a specific Intoxilyzer.  With accuracy verifications conducted as part of each individual test the COBRA system can be used to monitor the readings over time and address issues before they become problems.  Completed tests vs. tests with non-numeric results can be assessed as can incidences of non-numeric results vs. refusals with at least one breath sample.Demographic data can be collected and assessed along with the alcohol testing. Pretty charts and graphs can be generated for individual agencies showing the numbers of their arrests and time of day, age of arrestee and/or gender.  The individual officers can be assessed for the number of tests or arrests they are involved in and the alcohol concentrations of the subjects being tested. The refusal rate per department can be compared to the refusal rate statewide or the refusal rate of the individual officer. The downloaded data can also be used for assessing the training needs of the operators – both individual operators as well as statewide operators. If an operator makes frequent typographical errors that are hand-corrected later it could be the operator doesn’t know how to correct it during the data entry mode of testing. If large numbers of officers are running tests, without a new observation, after a mouth alcohol is detected the instructors and program managers will need to consider adding more information or revising how the training is currently being conducted. The DOFS will be able to generate control charts for the individual machines using the alcohol control tests (the dry gas samples run with each subject test) or they could program in weekly (remote) checks and downloads.

Remote checks meaning the DOFS can run tests such as diagnostics, DVM checks and control sample tests (simulator vapors or dry gas, whichever is connected) using the phone lines or high-speed connections. Remote troubleshooting becomes a reality. Updating driver’s records with breath-alcohol results and sending off necessary documentation to the courts and hearing officers can also be handled if the DOFS writes these functions into the programming. The COBRA system could be used to upload information to a virtual library for access by attorneys and others. Maintenances, inspections and repairs would be part of the database. Every test, in theory, should be there.  The availability of the database information to the defense community is an issue that has been fought time and again across the country and continues today. States that do not feel the breath alcohol program has anything to hide will post the database and certification pages for use in trial (Washington and South Carolina) while other states, like North Carolina, will fight sharing any of the data.

One of the most contentious issues in DUI defense over the past five years has been whether, and to what extent, the source code must be made available to the defense. The source code is that form of the software that can be read by a human programmer, as well as able to be processed by special programs that convert the source code into machine code. The full complement of source code is all of the electronic materials required to convert from the source code to the machine code. The conversion from source code to machine code is a one way process, that is, the machine code can always be reconstructed from the source code, but the source code cannot be reconstructed from the machine code. There are utilities that will convert machine code into a form that resembles source code. These utilities are called disassemblers and decompilers. While they generate text files that resemble the original source code, the text files generated in this fashion are not considered suitable for analyzing what the software does. Text files generated in this fashion are devoid of meaningful labels and comments, which were present in the original source code, and provide context for why the software operates the way it does. These comments and context are removed because they are not used by the computer when step by step instructions are created to instruct the computer. The source code is so named because all of the machine code can be traced to material present in the source code. One of the requirements imposed by DOFS requires CMI to disclose – on a very limited basis- its source code. A protective order and non-disclosure agreement acceptable to CMI must be in place. Furthermore, viewing of the source code must be done at CMI’s location. Why a non-disclosure agreement and protective order are necessary is difficult to explain – so long as you don’t think that hiding the source code is a legitimate exercise. The machine is patented, so there are no trade secrets involved. And requiring that the defense expert physically appear in Owensboro, Kentucky accomplishes little except making it more difficult and expensive for a defendant to obtain the source code. After all, one of these source codes can be sent halfway around the world in a zip file via e-mail.

In conclusion, it appears as if there will be an overall improvement in Georgia’s breath testing program. A control test with every subject test, downloading data, printing breath volumes, and reducing the number of potential interfering substances are all welcome developments. However, the refusal to adjust readings to reflect exhaled breath temperatures reflects a certain stubborn blindness, and the refusal to make the source code readily available perpetuates the tradition of secrecy cloaking breath alcohol testing. We had hoped for more transparency, but we are not surprised by what is coming our way soon.
Written by Allen Trapp who is board certified by the National College for DUI Defense and the author of Georgia DUI Survival Guide Visit Website

Written by Allen Trapp who is board certified by the National College for DUI Defense and the author of Georgia DUI Survival Guide Visit Website

You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

AddThis Social Bookmark Button

Leave a Reply

You must be logged in to post a comment.