Artículos del IFPUG

• 

  Schofield-Part1 > Counting Lines Of Code

  Tamaño:

  38 KB

  For the past four decades the nebulously-depicted line of code has been used to describe the size of a software project and often used as a basis for estimating schedule and resource needs.  Concurrently, software projects are noted for cost and schedule overruns, and often, for poor quality.

  Data from a series of Personal Software Process^SM courses provides the foundation for these three articles.  Because the requirements, instructor, and the lines of code counting specification for these programs were the same, the 50 sets of nine programs offers an extraordinary opportunity for comparing significant variation in software sizes for identical requirements by programmer.  Given the variation, often greater than an order of magnitude, for identical requirements, the use of lines of code as an indicator of software size is challenged.  Newly summarized analysis and resulting conclusions re-ignite this discussion, and in subsequent articles, an argument for the use of the data for measuring people is suggested.
• 

  Schofield-Part2 > Weakest Link

  Tamaño:

  46 KB

  Building on the data from Counting Lines of Code: Virtually Worthless for Estimating and Software Sizing this article revisits the assertion in The Statistically Unreliable Nature of Lines of Code [1] as to whether the programmer with the largest number of longest programs in a set of nine, in his / her peer class would be a candidate for retraining or some other more suitable line of work.  The analysis reinforces previous conclusions that individual software programmer performance is tied directly to the programming task at hand; previous work is not a reliable predictor of size, effort, or cost.  Nonetheless, some programmer practices might be worth emulating, others worthy of monitoring.
• 

  Schofield-Part3 > Lines Of Code Value

  Tamaño:

  161 KB

  The final article in this three-part series uses a uniquely consolidated version of the class data to explore two approaches for using lines of code data to identify "under performers."  In addition, several reasons are offered that substantiate how "under performers" deplete the organization of its value and capacity.  Strategies for reviving "under performers" is left to the discretion and imagination of the offended party.

   
• 

  Function Points to Measure Reusable Software - Spanish.pdf

  Tamaño:

  281 KB

  Este White Paper describe un enfoque del uso del Análisis de Puntos Función (FPA) en la identificación y medición de la reutilización del software. “Reutilización” se define como “utilizar los componentes de software existentes, en el desarrollo o mantenimiento de aplicaciones”. Existen otras formas de reutilización, tales como pruebas de regresión, plantillas de requisitos, diseño de plantillas, y así sucesivamente; sin embargo, estas formas de reutilización quedan fuera del alcance de este documento. En este documento no se tratará la característica general del sistema (GSC) #10 con respecto a la “reutilización” – que describe el grado en el que la aplicación y el código de la aplicación han sido específicamente diseñados, desarrollados y mantenidos para ser utilizables en otras aplicaciones –.