Page 148 - Automated Fingerprint Identification Systems (AFIS)

CURRENT ISSUES 133

to the source files at the company’s headquarters. This process allows for better
control of the application software, but it also is a one-size-fits-all approach. This
approach also requires that local software engineers become more dependent
on the software engineers at the company’s headquarters, who may be many
time zones away and therefore not easy to reach.
This leads to the issue of who really knows the AFIS system. As a defensive
measure, large AFIS agencies retain staff whose function is to understand the
workings of the system as well as the vendor and to identify problems and res-
olutions in partnership with the vendor. Unfortunately, many of these agencies
are losing the skilled staff who were part of the original AFIS installation team,
who are leaving because of retirement, opportunities in other fields, or other
reasons. Who will replace current AFIS management teams when they retire is
a growing concern.
In the past, it was not uncommon for an employee to start as a fingerprint
clerk, then be promoted to classifier or examiner. Along the way there may have
been opportunities for supervisory responsibilities. The craft master level might
be a position of an administrator in the identification agency. Good classifiers
and examiners might also have furthered their skills as latent print examiners.
In other words, there was a training ground for future administrators. That
training ground, however, is evaporating.
Learning the complexities of the Henry Classification System is no longer
necessary. There is little need to study ridge flows or tracings, or to argue over
the second level pattern. Examiners no longer handle a stack of inked tenprint
cards. Instead, they look at the images on their computer monitor and assume
that the information presented is correct. The systems have developed so
well that managers who do not have a fingerprint or systems background can
assume that the AFIS system is working well simply because it is working. That
is a very dangerous assumption.
Prior to 2000, there was a desperate need for COBOL programmers, most
of whom had retired, to correct outdated computer programs (the Y2K
problem). COBOL was so little used at that point that only a handful of people
knew how to program with it. Agencies could find themselves in a similar situ-
ation with AFIS as the skill sets or design and testing are lost. This leaves the
agency virtually at the mercy of vendors and their pricing structure.
The loss of tenprint examiners is second to the diminishing ranks of latent
print examiners. Latent print examiners have a unique set of skills. The most
important is their training and experience in making a latent print identifi-
cation. The examiner compares a partial print with candidate images from
tenprint records; this skill borders on the amazing. In addition, latent print
examiners have to know how to use AFIS to produce the best candidate list that
will contain the target. They may have to do this with no information other