The first result is in favour of open source development. Traditional developers might fear that open source has a high probability of producing unreadable code, of low quality and impossible to maintain. They may think that open projects manage to survive only because a large number of programmers, with infinite patience because of their personal interest, are available to correct bugs and provide add-ons.From the data we have examined,it seems that this conjecture can not be supported. The average percentage of acceptable components across the programs is still high, as half of the components are in good shape. On the other hand, the average percentage of components that must be rewritten is not prohibitive for a corrective activity on the code. However, it should be noted that according to Pareto’s law, a small percentage of the software will be responsible for the majority of the problems, so 5–6% of components needing to be rewritten is still a worrying result.
On the contrary, the second finding is against the open source development philosophy. Given the direct link between internal and external quality characteristics and the findings of this case study, it seems that the open source community should seriously take into account the need to develop higher quality code. This is suggested by the fact that, on average, almost half of the components of each application examined have not received the ACCEPT recommendation and must be reworked or revisited in some way (i.e. rewritten, tested, inspected or commented). Although the open source strength stems from the massive code-level peer review, such a suggestion implies that the way the code is structured necessitates even further work.Anyway, the quality of the code developed for Linux applications did not meet the requirements of the industrial standard we have considered, in contrast to what open source proponents have claimed up to now.
The first result is in favour of open source development. Traditional developers might fear that open source has a high probability of producing unreadable code, of low quality and impossible to maintain. They may think that open projects manage to survive only because a large number of programmers, with infinite patience because of their personal interest, are available to correct bugs and provide add-ons.From the data we have examined,it seems that this conjecture can not be supported. The average percentage of acceptable components across the programs is still high, as half of the components are in good shape. On the other hand, the average percentage of components that must be rewritten is not prohibitive for a corrective activity on the code. However, it should be noted that according to Pareto’s law, a small percentage of the software will be responsible for the majority of the problems, so 5–6% of components needing to be rewritten is still a worrying result. On the contrary, the second finding is against the open source development philosophy. Given the direct link between internal and external quality characteristics and the findings of this case study, it seems that the open source community should seriously take into account the need to develop higher quality code. This is suggested by the fact that, on average, almost half of the components of each application examined have not received the ACCEPT recommendation and must be reworked or revisited in some way (i.e. rewritten, tested, inspected or commented). Although the open source strength stems from the massive code-level peer review, such a suggestion implies that the way the code is structured necessitates even further work.Anyway, the quality of the code developed for Linux applications did not meet the requirements of the industrial standard we have considered, in contrast to what open source proponents have claimed up to now.
การแปล กรุณารอสักครู่..
