Gun Database Blues

America's law enforcement will resort to any junk science imaginable to incarcerate would be felons and misfits. The following method of collecting evidence is as wrong headed as collecting DNA from identical twins. Plain and simple it doesn't work and will send many innocent people to jail. Residents of greater Washington, after weeks of siege, could finally stand up straight when police apparently closed in on the infamous sniper. The public-event cancellations and school lock downs had real effects on nearly everyone's lives and psyches. Predictably, some politicians exploited the episode for partisan gain. Some called for establishing a national ballistics database, suggesting anyone who questioned the idea must be a National Rifle Association nut. I'm not going to jump into the gun control debate, thank you. But I am concerned whenever, pardon me, magic-bullet technology solutions are bandied about by politicians. DATABASES SUCH AS that proposed in HR 408, the Gun Ballistics Safety Act of 2001, contain ballistics records from firing each new gun before it is sold. Even cursory research into so-called gun fingerprinting casts doubt on its usefulness. Among the problems: * Firearm ballistics characteristics change with repeated firings, and can be altered deliberately. * Many crimes are committed with old guns that wouldn't be in a database, or with stolen ones for which a trace would lead nowhere. * Numerous guns — say, a particular manufacturing run from one supplier or similarly designed firearms by different manufacturers — can have such similar ballistics that database queries could overwhelm forensics investigators with the number of hits. To be sure, several states have ballistics databases, and officials say using them has led to arrests. But several published studies have called such databases inaccurate or impractical. Now the Bush administration has wisely asked the Bureau of Alcohol, Tobacco and Firearms to do a feasibility study. National databases of anything have proven notoriously expensive and difficult to implement. Can anyone remember the FBI's travails in the early 1990s with its national database of human fingerprints? Policy made under fire is often bad policy, but agency IT and program people, not the politicians, would be blamed for the inevitable problems. This one needs more study.