Beginning in the late 1980s in Washington State, and more recently and frequently in other jurisdictions including Arizona, Connecticut, Florida, Georgia, Kentucky, Louisiana, Massachusetts, Minnesota, New Jersey, New York, Tennessee, Texas, and the Province of Ontario, Canada, defendants in driving-under-the-influence-of-alcohol (DUI) cases have made discovery demands for the Source Code∗ of the computer software of the evidential breath alcohol analyzers with which they had been tested.

The Committee on Alcohol and Other Drugs has been advised that the great majority of trial
and appellate court decisions on such Source Code issues to date have denied the defense
access to the Source Code, chiefly on the grounds that it is not material or necessary to the
challenge of the evidential breath-alcohol analyzer and/or the results yielded by that device,
or that the state has no duty to disclose information not in its control, custody, or possession.
However, in aggregate, Source Code litigation has resulted in great expenditures of effort,
time, and costs as well as substantial delays in resolving the underlying cases, without useful
outcomes. Therefore, the Committee has in the public interest examined in detail the Source
Code litigation discovery issue, from a forensic science and technological perspective.
It is the position of the National Safety Council Committee on Alcohol and
Other Drugs that access to the Source Code of the software of an evidential
breath-alcohol analyzer is not pertinent, required, or useful for examination
or evaluation of the analyzer’s accuracy, scientific reliability, forensic
validity, or other relevant characteristics, or of the trustworthiness and
reliability of analysis results produced by the analyzer. These matters can be
and have been fully assessed and examined by multiple other well established
and recognized methods and procedures in common use worldwide; and
many other adequate and appropriate means exist to challenge evidential
breath-alcohol analysis results.
∗
Source Code is a series of statements written in human-readable computer programming
language. It is the basis of all computer programs, and consists of a structured set of
instructions that cause a computer or microprocessor-equipped device to perform specific
operations.
National Safety Council's
Committee on Alcohol and Other Drugs
2/3
COMMENT
The above position of the Committee is reflected universally in operational practices in
evidential breath-alcohol testing, nationally and internationally. The comprehensive national
program for evaluating evidential breath-alcohol analyzers for conformity with Federally
established Model Specifications for Evidential Breath Testing Devices, an activity carried
out by the National Highway Traffic Safety Administration of the U. S. Department of
Transportation to establish and maintain a Conforming Products List of such devices, does
not use or evaluate Source Codes of the analyzers in that process. Likewise, the International
Recommendation on Breath Alcohol Analyzers, OIML R 126 “Evidential breath analyzers,”
promulgated by the International Organization of Legal Metrology (OIML), an international
treaty organization of which the United States of America is a Member State, does not refer
to or include the Source Code of analyzers in its Recommendation, the purpose of which is to
define the performance requirements of evidential breath analyzers for alcohol and the means
and methods employed in testing them. The evidential breath-alcohol analyzer instrument
standards and approval procedures officially used in Canada have been recommended by the
Alcohol Test Committee of the Canadian Society of Forensic Science. These standards and
procedures do not mention or utilize analyzer Source Codes. The leading treatises on the
subject of quality assurance in breath-alcohol analysis, published in the peer-reviewed
scientific literature, authored by Dubowski (1994) and Gullberg (2000), are intended to
enhance the reliability and validity of evidential breath-alcohol testing practice. They do not
mention Source Codes, although both scientists are fully familiar with that aspect of
computerization.
The References cited below confirm and support the Committee’s foregoing position.
REFERENCES
1. “A Guide to Type Approval Procedures for Evidential Breath Alcohol Testing
Instruments Used for Road Traffic Law Enforcement in Great Britain.” Home Office
and Forensic Science Service. London, HMSO Publications, (Oct. 1994).
2. California v. Trombetta, [U. S. Supreme Court] 467 U. S. 479 (1984).
3. Dubowski, K. M.: “Quality Assurance in Breath-Alcohol Analysis,” J. Analyt.
Toxicol. 18: 306-311 (Oct. 1994).
4. Federal Rules of Evidence. ARTICLE IV, Rule 402 (2008).
5. Federal Rules of Evidence. ARTICLE IV, Rule 403 (2008).
6. Gullberg, R. G.: “Methodology and Quality Assurance in Forensic Breath Alcohol
Analysis,” Forensic Science Review 12: 49-67 (Jan. 2000).
7. Hodgson, B. T.: “The Validity of Evidential Breath Alcohol Testing,” [Review Article]
Can. Soc. Forensic Sci. J. 41: 83-96 (2008).
8. King, M. P. (Special Master): “SUPPLEMENTAL FINDINGS AND CONCLUSIONS
OF REMAND COURT,” submitted to the New Jersey Supreme Court
Nov. 8, 2007. [Supreme Court of New Jersey, September Term 2005, Docket No.
58,879, State of New Jersey v. Chun, et al.]
3/3
[The New Jersey Supreme Court Special Master’s “Supplemental Findings
and Conclusions of Remand Court,” on the Source Code issue and dated Nov.
8, 2007, in State of New Jersey v. Chun, et al., contain much information
about the complexities of examining and evaluating Source Codes of the
software of evidential breath-alcohol analyzers; and establish that these are
not simple or routine tasks. The Special Master, at pp. 51-52, recorded that an
expert witness for the defendants, without rebuttal, testified that “the
Alcotest’s software was ‘far too complex’ to test” and that the same witness
also “…estimated that it would require all of mankind for the rest of time to
test all of the paths in the Alcotest’s source code.”]
9. Koehn, M.: REPORT – Best Practices: “Producing Source Code in Litigation,” Digital
Discovery & E-Guidance 8 (No. 3) 1-3: (March 1, 2008) [BNA Report #ISSN 1537-
5099].
10. McShane, S. F., and Hammer, I. J.: Intellectual Property: “Protecting Source Code,”
New Jersey Law J. 181: 231-233 (July 18, 2005).
11. National Highway Traffic Safety Administration, Highway Safety Programs: “Model
Specifications for Devices to Measure Breath Alcohol,” Federal Register 58: 48705-
48710 (Sept. 17, 1993).
12. OIML: “Recommendation R 126 – Evidential breath analyzers.” Paris, International
Organization of Legal Metrology (1998).
13. “Recommended Standards and Procedures of the Canadian Society of Forensic Science
Alcohol Test Committee,” Can. Soc. Forensic Sci. J. 36: 101-127 (2003).
14. Short, C.: “Guilt by Machine. The Problem of Source Code Discovery in Florida DUI
Prosecutions,” Florida Law Review 61: 177-201 (Jan. 2009).
15. State of New Jersey v. Chun, et al. [New Jersey Supreme Court] 191 N. J. 308 (2007),
194 N. J. 54 (2008).
16. Strutin, K.: “Examining Source Code Evidence,” New York Law Journal, pp. 5-8
(November 14, 2007).

Adopted February 16, 2009 by the NSC CAOD

This concludes the position of one non-judicial government agency deciding what they expediently think is vs. what is not relevant in a criminal DUI trial.