Size of the team and project often determines the Death Match Project's chances of failure, says Dr R Srinivasan

" The odds of successful completion may drop considerably in medium size projects and may become almost impossible in large-scale projects"

CONTINUING The discussion we had in the last issue on Death March Projects, Yourdon provides another way to characterize the it, "A Death March Project is one for which an unbiased, object risk assessment (which includes an assessment of technical risks, personal risks, legal risks, political risks, etc.) determines that the likelihood of failure is greater than 50 percent." If we try to take stock of some of the projects under this description it will be possible to classify them under four simple categories depending on the size of the project team and the duration of the project. A project having a team of less then ten members and duration of three to six months will to the small-scale category. In such a project the probability of coming out of problems and bringing the project to a successful end is high because of the good coherence among the team members, particularly if the team has a set of highly motivated people.

However, Yourdon brings in a rider here that this will happen as long as the commitment is not more than six months or so. On the other hand the second category project having twenty to thirty members with a schedule of one or two years, is called medium sized. Those having more than 100 people with a project schedule of three or four years belong to the large-scale category. The odds of successful completion may drop considerably in medium size projects and may become almost impossible in large-scale projects. The fourth category, according to Yourdon, are mind boggling projects like that of NASA's project of landing a person on the moon or like the missile program of the Government of India where, in addition to the large number of developers from different scientific organization, there will be groups from private and public sector industries, consultants, etc,. An excellent co-ordination is needed in such projects. In addition to classifying the projects according to the size and duration it will be possible to characterise the degree of Death March Projects depending upon the number of end users. Larger the number of end users, queries and classifications will be more and will increase in a disproportion ate manner during the project maintenance or customer support.

In one of the earlier articles I have mentioned that it is the Project Manager's responsibility to see that the project is completed successfully. As leaders in the IT industry, Project Managers play a vital role right from deciding whether the organisation and the members have the expertise to undertake a project and then to orchestrate carefully designed plans to complete project successfully, often incorporating complex dynamic and changing requirements. Thus project management today is very similar to the procedures and process set by people in the early days. Richard Murch in his book on, 'Project Management - best practices for IT professionals', cites examples of such people. A typical one is that of Frederick Taylor who wrote a book in 1911 on, 'Principles of Scientific Management'. Even though he was working in steel mills, his management principals exactly depict what we have today in the management of software projects, particularly in the efforts put in to achieve the best productivity of team members.

Taylor's management principals are summarized as: 1) Analyse each job to specify optimal procedures; 2) Match skills with tasks to be accomplished; 3) Understand worker characteristics that are important for increased productivity; 4) Train workers to be more productive; 5) Set " fair day's work" standard for productivity expectation; 6) Document worker performance; 7) Complete management and reporting of all work. Another example is the origin of Gantt chart that we use today in Project Management. It was the discovery of Henry Gantt who was an associate of Frederick Taylor. It is in fact inscribed on Taylor's gravestone in Pennsy1vania, "Father of Management". Dr Edwards Deming is one more stalwart who introduced and stressed the importance of quality in product development and aptly he has been acclaimed to be the "Father of Quality". Later on we have been seeing in recent times many gurus and leaders in the areas of Software Project Planning, Execution and Control through International Standards. While we have the guidelines, methods and standard processes available, why should we get into Death March Project? Yourdon, out of his experience in visiting several software development centers across the world, comes up with many reasons in his book. They are: 1). Politics, 2). Naïve Promise made by marketing and senior executives and naïve project managers, 3). Naïve optimism of youth saying, "We can do it over the weekend', etc.

The most important point to be borne in mind before agreeing to undertake a software development project is viable to be done by the organisation and do they have the right expertise.

(The author is Chief Technology Officer, Internet Component Management Group, Bangalore and can be contacted at: r.srinivasan@iCMGworld.com)