Software is used in every industry; the cars we drive and mobile we use rely heavily on software. Companies from varies industries has an annual spending on software projects either to develop new software or maintain existing software licenses, computer hardware or any IT related spending. Charette stated that depending on the business industry the IT spending ranges from 4 – 10 % of its revenue (Charette, 2005), being a government, non-profit organization, financial institution or a retailer each of these industries require spending in IT projects either to introduce new services, products or facilitate its operations. In this article I will analyze the trend behind software failures, and if it is related to a specific industry, or not.
Software failure implications vary based on the criticality of the software and industry, it could lead to waste of billions of dollars, or result in the death of people. As well we could notice that software failure is not limited to an industry as each industry suffered from software failures. People have learned from their mistakes and developed methodologies to manage software development risks, and to minimize the impact and frequency of software failures, in addition to that good project management is needed to achieve better success rate. Based on a report by IBM done to identify success rate trend, it appeared that software project success rate in getting better by time taking into consideration the increased complexity. However, the success rate is increasing slowly, for example the software project success rate in 2001 was 28% while in 2003 its 31% (Marasco, 2006).
Reasons behind software failures are plenty, starting from unrealistic scope, improper identification of requirements, poor project management and a lot of other reasons. I agree with Charette in his finding that rarely software projects fail for one reason only; usually it’s a mixture of reasons that cause software projects to fail and software to fail as well. Hamill and Goseva-Popstojanova in 2009 in their research found that software fail regardless of their industry, development methodologies applied and coding language used; and they fail during implementation phase. And this finding supports the idea that software fails in almost every industry and its not industry related, the translation from user requirements into developers language and the proper understanding of user requirements is very critical to project success.
In conclusion, software failures are not industry related. Software engineers are learning from software failures are developing software methodologies to overcome known limitations and reduce software failure probability, the impact of financial crisis and organization’s tendency to lower cost may have affected and increase software failure rate, this because organizations are expecting more for less.
Reference:
Charette, R.N., (2005), ‘Why Software fails [software failure]’, IEEE Journals & Magazines, [Online]. Vol 42, Issue 9, pp. 42-49. Available from: http://ieeexplore.ieee.org.ezproxy.liv.ac.uk/xpls/abs_all.jsp?arnumber=1502528&tag=1 (Accessed 5 May 2012)
Hamill, K. & Goseva, K. 2009, "Common Trends in Software Fault and Failure Data", IEEE Transactions on Software Engineering, [Online], vol. 35, no. 4, pp. 484-496 [Online]. Available from:http://ieeexplore.ieee.org.ezproxy.liv.ac.uk/xpls/abs_all.jsp?arnumber=4760152&tag=1 (Accessed: 05 May 2012)
Marasco, (2006), ‘Software development productivity and project success rates: Are we attacking the right problem?’, [Online]. Available from: http://www.ibm.com/developerworks/rational/library/feb06/marasco/
(Accessed: 05 May 2012)
No comments:
Post a Comment