Backfiring: converting lines of code to function points

The availability of empirical data from projects that use both function-point and lines-of-code metrics has led to a useful technique called backfiring. Backfiring is the direct mathematical conversion of LOC data into equivalent function-point data. Because the backfiring equations are bidirectional, they also provide a powerful way of sizing, or predicting, source-code volume for any known programming language or combination of languages. The function-point metric, invented by A.J. Albrecht of IBM in the middle 1970s, is a synthetic metric derived by a weighted formula that includes five elements: inputs, outputs, logical files, inquiries, and interfaces. IBM put it into the public domain in 1979, and its use spread rapidly, particularly after the formation of the International Function Point Users Group (IFPUG) in the mid-1980s. By then, hundreds of software projects had been measured using both function points and lines of source code. Since an application's function-point total is independent of the source code, this dual analysis has led to several important discoveries.<
>