Glossary


Process Feedback

A-B-C- D-E-F- G-H-I- J-K-L- M-N-O- P-Q-R- S-T-U- V-W-X- Y-Z

Definitions/Uses

1996
Reference /Lehman, 1996a/ Laws of Software Evolution Revisited

Definition/
Use 8th law of software evolution: Feedback System: E-Type evolution process are multi-level, multi-loop, multi-agent feedback systems and must, in general, be treated as such to achieve major process improvement.
Positive feedbacks triggers or accelerate growth may lead to instability. Negative feedback, on the other hand, has a stabilising influence. When negative feedback is applied over some forward path element, be it a single mechanism or a subsystem, changes in the output of that element in response to changes in its characteristics are reduced by aproximately the gain in the feedback loop. The precise impact depends, of course, also on the delay or phase shift in the loop. One achieves global stability in the face of changes in element characteristics.

1994
Reference /Garg, 1994/ The SMART Approach for Software Process Engineering

Definition/ Uses An important aspect of our process engineering life-cycle is that it does not assume that once a process model has been developed it remains fixed forever. On the contrary, we anticipate a process engineering life-cycle in which the process models can evolve to accommodate changes in the execution of the process. An example change in the environment could be when a new way of doing a particular aspect of the process is discovered. For instance, suppose that programmer modifies code by using an edit-compile debug cycle, and that this has been modeled in the embedding knowledge-base. Therefore, whenever there is a task of modifying code, it is embedded within the PSEE with the activities of edit, compile, and debug. At some point during the execution of these activities, a programmer may discover a static analysis tool and start using it with the edit activity. Therefore, the task of modifying code becomes into an edit-analyze-compile-debug cycle. The new activity of analyzing the code can be recognized by the PSEE, and an appropriate message can be sent to the process embedding tool.
The feedback can be utilized by the process embedding tool to improve its transformation of future process models. Therefore, future programmers need not discover the use of the static analysis tool themselves, since knowledge of its use will be available for them from past experience with the process embedding tool.
In this manner, the process life-cycle is an evolutionary life-cycle in which the processes developed are incrementally enhanced and continuously improved.

1993
Reference /Lonchamp, 1993/ A Structured Conceptual and Terminological Framework for Software Process Engineering

Definition/
Use The carrying back of some of the effects of a model driven process.

1996
Reference	/Lehman, 1996a/ Laws of Software Evolution Revisited
Definition/ Use	8th law of software evolution: Feedback System: E-Type evolution process are multi-level, multi-loop, multi-agent feedback systems and must, in general, be treated as such to achieve major process improvement. Positive feedbacks triggers or accelerate growth may lead to instability. Negative feedback, on the other hand, has a stabilising influence. When negative feedback is applied over some forward path element, be it a single mechanism or a subsystem, changes in the output of that element in response to changes in its characteristics are reduced by aproximately the gain in the feedback loop. The precise impact depends, of course, also on the delay or phase shift in the loop. One achieves global stability in the face of changes in element characteristics.
1994
Reference	/Garg, 1994/ The SMART Approach for Software Process Engineering
Definition/ Uses	An important aspect of our process engineering life-cycle is that it does not assume that once a process model has been developed it remains fixed forever. On the contrary, we anticipate a process engineering life-cycle in which the process models can evolve to accommodate changes in the execution of the process. An example change in the environment could be when a new way of doing a particular aspect of the process is discovered. For instance, suppose that programmer modifies code by using an edit-compile debug cycle, and that this has been modeled in the embedding knowledge-base. Therefore, whenever there is a task of modifying code, it is embedded within the PSEE with the activities of edit, compile, and debug. At some point during the execution of these activities, a programmer may discover a static analysis tool and start using it with the edit activity. Therefore, the task of modifying code becomes into an edit-analyze-compile-debug cycle. The new activity of analyzing the code can be recognized by the PSEE, and an appropriate message can be sent to the process embedding tool. The feedback can be utilized by the process embedding tool to improve its transformation of future process models. Therefore, future programmers need not discover the use of the static analysis tool themselves, since knowledge of its use will be available for them from past experience with the process embedding tool. In this manner, the process life-cycle is an evolutionary life-cycle in which the processes developed are incrementally enhanced and continuously improved.
1993
Reference	/Lonchamp, 1993/ A Structured Conceptual and Terminological Framework for Software Process Engineering
Definition/ Use	The carrying back of some of the effects of a model driven process.

Identification

Also as

Definitions/Uses

Related terms in the glossary

Publications on this area

Projects on this area