Sommerville-Software-Engineering-10ed

24.5 ■ Software measurement 725
On the other hand, the process changes might lead to many new, happy customers
who wish to participate in the product development process. They therefore generate
more change requests. Changes to the process of handling change requests may contribute
to this increase. If the company is more responsive to customers, they may
generate more change requests because they know that these requests will be taken
seriously. They believe that their suggestions will probably be incorporated in later
versions of the software. Alternatively, the number of change requests might have
increased because the beta-test sites were not typical of most usage of the program.
To analyze the change request data, you do not simply need to know the number
of change requests. You need to know who made the request, how the software is
used, and why the request was made. You also need information about external factors
such as modifications to the change request procedure or market changes that
might have an effect. With this information, you are in a better position to find out if
the process changes have been effective in increasing product quality.
This illustrates the difficulties of understanding the effects of changes. The “scientific”
approach to this problem is to reduce the number of factors that might affect
the measurements made. However, processes and products that are being measured
are not insulated from their environment. The business environment is constantly
changing, and it is impossible to avoid changes to work practice just because they
may make comparisons of data invalid. As such, quantitative data about human
activities cannot always be taken at face value. The reasons a measured value
changes are often ambiguous. These reasons must be investigated in detail before
any conclusions can be drawn from any measurements.
24.5.4 Software analytics
Over the past few years, the notion of “big data analysis” has emerged as a means of
discovering insights by automatically mining and analyzing very large volumes of
automatically collected data. It is possible to discover relationships between data items
that could not be found by manual data analysis and modeling. Software analytics is
the application of such techniques to data about software and software processes.
Two factors have made software analytics possible:
1. The automated collection of user data by software product companies when their
product is used. If the software fails, information about the failure and the state of
the system can be sent over the Internet from the user’s computer to servers run by
the product developer. As a result, large volumes of data about individual products
such as Internet Explorer or Photoshop have become available for analysis.
2. The use of open-source software available on platforms such as Sourceforge
and GitHub and open-source repositories of software engineering data (Menzies
and Zimmermann 2013). The source code of open-source software is available
for automated analysis and can sometimes be linked with data in the opensource
repository.