S. McConnell. Gauging software readiness with defect tracking. IEEE Software, 14(3), May/June 1997.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....avoid the problems of low quality, but run the risk of exceeding deadlines and being penalised for the late delivery. Therefore we need a method to predict when software is ready for release. Such estimations can be made by defect tracking. Figure 12 Viewpoint Directory Legend McConnell (McConnell, 97) describes four techniques to perform such estimations, Defect Density (Musa et al. 87) this measures the number of defects per KLOC 1 from previous projects. From this we can determine the expected number of defects in the current system, providing we know the program s length, and ....

MCCONNELL, S.: "Gauging Software Readiness With Defect Tracking", IEEE Softw., 1997, 14 (3) pp. 136. - 30 -

....all remaining defects. This is obviously infeasible for any commercial product. Even if resources are available, it will take a prohibitive amount of time to find all bugs in a large program [2] Sampling based methods have been suggested for estimating the number of remaining defects. McConnell [14] has given a method that involves using two independent testing activities, perhaps by two different teams of testers. This and other sampling technique assume that faults found have the same testability as faults not found. However, in actual practice, the faults not found represent faults that ....

....the same testability as faults not found. However, in actual practice, the faults not found represent faults that are harder to find [13] Thus such techniques are likely to yield an estimate of faults that are relatively easier to find and thus less than the true number. Fault seeding methods [14] suffer from similar problems. It is possible to estimate the defect density based on past experience using empirical models like the Rome Lab model [7] or the model proposed by Malaiya and Denton [9] The estimates obtained by such models can be very useful for initial planning, however these ....

S. McConnell. Gauging software readiness with defect tracking. IEEE Software, 14(3), May/June 1997.

....This is obviously infeasible for any commercial product. Even if resources are available, it will take a prohibitive amount of time to find all bugs in a large program (Butler and Finelli 1993) Sampling based methods have been suggested for estimating the number of remaining defects. McConnell (McConnell 1997) has given a method that involves using two independent testing activities, perhaps by two different teams of testers. This and other sampling techniques assume that faults found have the same testability as faults not found. However, in actual practice, the faults not found represent faults that ....

....as faults not found. However, in actual practice, the faults not found represent faults that are harder to find (Malaiya and Yang 1984) Thus, such techniques are likely to yield an estimate of faults that are relatively easier to find and thus less than the true number. Fault seeding methods (McConnell 1997) suffer from similar problems. It is possible to estimate the defect density based on past experience using empirical models like the Rome Lab model (Lakey and Neufelder 1997) or the model proposed by Malaiya and Denton (Malaiya and Denton 1997) The estimates obtained by such models can be very ....

McConnell, S. (1997, May/June). Gauging software readiness with defect tracking. IEEE Software 14(3), 135--136.

No context found.

S. McConnell. Gauging software readiness with defect tracking. IEEE Software, 14(3), May/June 1997.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC