At academic conferences on software measurement and analysis, I feel frustrated when I discover the software metric adopted by analysts. They tend to evaluate software development productivity using SLOC as a metric.
The reason for this is almost always the same: we need data, large amounts of data: the only available and reliable data are SLOC. Which is the correlation between SLOC and software development? The number of lines of code per day? Is this a real measure of productivity?
If you ask them why they don’t use Function Points instead of SLOC, you will get confused! Which type of Function Points? IFPUG, COSMIC, SNAP or other variants? When are they applicable?
Last time I heard the same old story I was at OSS 2013, the International Conference on Open Source Systems – this year focusing on validation.
I was there to give my speech at the end of the conference.
In the first part of my presentation, I gave some hints on open source, innovation and next challenges in the context of the “nexus of events”.
The second part of the presentation focused on quality gaining momentum in open source software. I want to demonstrate that you can assess trustworthiness and reliability of open source software using a PMAI (Plan-Measure-Assess-Improve) approach. As you can see, Spago4Q can be used in various contexts, such as giving open information about SpagoBI project, supporting the trustworthiness of information provided by open source communities (its adoption in OW2 SQuAT program is coming soon) and monitoring the entire Application Lifecycle Management process at Engineering Group.
The latter is a key point!
Engineering Group’s Software Laboratories bring together hundreds of technical managers, architects and developers supporting the worldwide development of software products and projects and application maintenance services. These activities are developed in compliance with customers’ and/or corporate procedures and with various quality systems such as ISO and CMMi.
To make this happens efficiently, a common infrastructure integrates various development tools and collects production information. Here is the Productivity Intelligence role: using business intelligence techniques to monitor and improve productivity, through integration of data coming from different sources.
Which information? No SLOC at all!
Productivity is not just lines of code per day or man-hours per day! It’s the sum of three dimensions of analysis: Technical (ex: software measures, bug fixes, test coverage), Economical (development costs and efficiency) and Social (customers’ and developers’ satisfaction). Productivity comes both from quantitative and qualitative data.
Thanks to the QEST nD model integrated in Spago4Q, this 3D evaluation allows users to take advantage of different views and capabilities to navigate (drill-down) information through the company organizational levels (top managers, technical managers, project managers), by means of both unified and single-dimension views.
Join us at IT Confidence 2013 and see Spago4Q with QEST nD in action!