Response to “Analysis eight secrets of the Software Measurement”

Introduction

This is a response to the article “Analysis eight secrets of the Software Measurement” posted by s1301111 [1].

In this post [1], the author divides his article into seven sections based on the eight secrets mentioned in his referenced article [2]. The author talks about the eight secrets of software measurement by summarizing the referenced article and describing his experience and opinions. I think it might be a little bit longer and lack of a main point, so I think some sections could be cut off. This post will mainly discuss one point he mentioned in his article which I think is the most important secret of software measurement, that is “Both establishing and keeping a measurement program are hard”. For other sections he mentioned in his article, I’ll choose some and briefly demonstrate my points.

“It’s not about the metrics”

In this section, I think the main point he wanted to convey is that the measurement is not about the metrics. So, I think what are metrics is not very important, so it’s not necessary to use such a long paragraph to explain “metrics” and even demonstrate six principles to establish good metrics. I agree with his main point and think that what our final goal for the measurement program is, but not just regard it as a tool.

“Success comes from channeling an organization’s pain into action”

The author describes mainly in two parts. The first is that a strong motivation is important for improving and understanding the measurement. Besides, understanding the meaning and difficulty of software measurement is also a vital step for achieving the plan. I agree with his point. A strong motivation can give us strength to take actions. However, we also need to consider the difficulty before we take actions in case the measurement will quickly be found inappropriate.

“Both establishing and keeping a measurement program are hard”

In this section, the author demonstrates his points that both establishing and keeping a measurement program are very important and difficult. I agree with him, and I’ll focus on this section. When dealing with the measurement program, we need to pay attention to establishing and keeping. None of these two steps could be ignored.

1. Establishing

For the establishing step, we need to take some factors into consideration before defining a measurement program. Building an empirical model is very necessary, which means we must identify the technical, cultural, organizational, and business issue [3]. These involve both technical and business goals. I think the technical goal is essential and indispensable. Sometimes we just do a project at school but will not apply it to the market, so technical goal is the most important thing we need to considerate. We need to analyze the procedure and find some resources to define the method. For the company, they always overlook the importance of technical goal and ignore the business goal. However, to successfully establish a measurement program, developers need to first consider the goals in the context of the organization’s underlying business strategy [3]. Because the measurement program may be forced to stop due to lack of funding or have no profit. I think it’s similar with starting an entrepreneurial enterprise. We may come up with lots of entrepreneurial ideas, such us a transnational corporation, or just a small retail business selling some hand-made products. However, before we start, we need to consider if we can make profit, how to find funding, who are the customers and so on. Business goal is a very important factor and it’s really difficult to start due to lots of complicated and unexpected factors.

2. Keeping

After successfully establishing the measurement program, it’s also difficult to implement and sustain. We need to review the measurement program regularly, to ensure if it operates appropriately, and gradual program optimization [3]. If it does not meet the process made in early plan, we need to find the reasons, improve the program, or revise the plan considering the current state.

Other than the technical measurement, business is also an important factor for company to measure if they could continue. We should keep track of it gradually, to avoid the great loss. When I was in the university, there was a special study room that they aimed to give students a comfortable study environment. Because our study room and library closed very early in the evening, students always have no place to study. This company found the opportunity and set up a study room. It provided high speed WIFI, coffee, quiet rooms, group study rooms and so on. Especially, it opened all day. It was really a perfect place for students. We all came there, enjoying the comfortable environment during study. However, we never imagined that it forced to close after one year with nearly fifty thousand pounds loss. Because it supplied such really good conditions, it cost too much money. However, it mainly faced to students who have no ability to afford too much fee for study room. So they can’t make the fee too high and the profit can’t meet their cost. I think it was really a good plan, however, they didn’t keep track of the business state gradually and revise their plan, and finally they had to close the study room when they started to notice the great loss.

Others

For other measurement secrets the author described in his article, I think he just agrees with the original article and doesn’t add something. As his article is a little bit long, I recommend that they could be cut off. To expend, I’ll now provide an additional secret of the software measurement which I think is important.

Measurement is being paid a lot attention from developers. However, it always be ignored by practitioners. However, the practitioners always trust empirical evidence of a measurement without its technical grounding. So, developers should communicate closely with them to understand the valid uses of a software measurement. [4] This gap should be paid more attention to.

Conclusion

In this article, I mainly discuss one key secret of software measurement that we should pay attention to the establishing and keeping a measurement program. In addition, I also give my opinion based on the author’s article that measurement is not about metrics and the motivation is very important, we also need to take it into actions. Finally, I provide an addition secret which I think is also very important that we need to bridge the gap between developers and practitioners.

Reference

[1] https://blog.inf.ed.ac.uk/sapm/2014/02/14/analysis-eight-secrets-of-the-software-measurement/

[2] http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=207238

[3] http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=582974

[4] http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=582973