Decision Oriented Model of Software Processes

Posted on April 22, 2011

0


 

Author:  Sanjay Goel, http://in.linkedin.com/in/sgoel

________________________________________

Toffolon and Dakhi [1]   have proposed a decision oriented model of software processes.  As per this view  software engineering is a nexus of interwined, concurrent, and mutually suppotive decisions (under uncertainties)  and operational activities. The decision making is carried out by many stakeholders with respect to four subspaces related to software projects:

(i) Problem space:  This space represents the customer’s business where the customer’s/ users’ problems and their organizational solutions are defined.

(ii) Solution space:   This space represents the architect’s business where the computer solutions of the customer/user’s problems are defined.

(iii) Construction space: This space represents the developer’s business where these solutions are implemented.

(iv) Operation space: This space represents the end user’s business where the software usability  from the user’s perspective as well as its contribution to the organization’s competitiveness are evaluated.

The decisions in these four spaces are driven by two broad categories of purposes as suggested by  Brooks in his seminal paper [2]:

(i) decisions to manage complexity and risk, the two essential characteristics of software, and

(ii) decisions to reduce the negative impacts of two kinds of accidental characteristics of software, i.e., uncertainty and complications.

The essential chracterstics are inherent in the nature of the software whereas the accidental factors are extrinsic and therefore  not inevitable.   Complexity is a problem of system interrelations. It  has a coherent architecture and can be effectively managed. In contrast, complications  are largely random and hence become unmanageable over time [3]. Something can be complex without being complicated, or complex because it is complicated. Complications arise because of poor management, inefficient and inflexible processes, lack of skilled people, lack of timely and accurate information, inefficient decion making environment [4]. Imperfectly understood, unpredictable interrelations create complications.

References:

[1].  Claudine Toffolon, Salem Dakhli, A decision oriented model of software engineering processes, Proceedings European and Mediterranean Conference on Information Systems 2007 (EMCIS2007),  Polytechnic University of Valencia, June 24-26 2007, http://www.iseing.org/emcis/EMCIS2007/emcis07cd/EMCIS07-PDFs/702.pdf.

[2] F. P. Brooks (1987), No silver Bullet:-Essence and Accidents of Software engineering, IEEE 

[3]  http://www.cioupdate.com/insights/article.php/11049_3866441_2/Reinventing-IT-Project-Management—Peter-Drucker-Style.htm

[4] http://www.industryweek.com/articles/mastering_complexity_driving_out_complication_23844.aspx?SectionID=3

[5]  Taxonomy of Decision Making wrt Software Development

Advertisements
Posted in: Uncategorized