Chinese plan to improve the software industry
Recently the Nesma office was contacted by a representative of China System and Software Process Improvement Association (SPI中国) with the question if Nesma would be willing to give a keynote speech at the annual conference on software estimation. In the same time, SPI China also reached out to the International Software Benchmarking Standards Group (ISBSG) to discuss and renew a partnership that had faded somewhat the last couple of years. Being Nesma board member and ISBSG president, obviously I was curious what this sudden interest out of China was about. After exchanging some e-mails, I understood that Software Cost Estimation is becoming a very important part of the Chinese government plan to improve the Chinese software industry maturity. 在 2017, the new 5-year plan will take effect and all preparations are now done in order to build up the right knowledge and to start the right partnerships. So of course, I was absolutely thrilled to visit SPI China and to discuss the possibilities to collaborate with Nesma and ISBSG in order to help the Chinese industry mature.
My first visit to China was in 2008 when the annual ISBSG was organized in Nanjing. Now I went for the second time, 在四月份 2016, this time to Beijing and it turned out that actually some of the same people were involved as back then in Nanjing. When Mr. Richard Wong, Secretary General of SPI China was introducing the SPI China organization to me in a powerpoint, there were some pictures of important events in the past. And suddenly I saw myself there… in a picture taken with the ISBSG group taken at the workshop in Nanjing. That was a funny surprise! Next to the partnership discussions, I also gave a training in Nesma functional size measurement and software estimation to a number of SPI China consultants. A few of them already had quite some experience in functional size measurement and I was excited to see that they were all very interested in the Nesma method and the associated guides (especially the guideline for the use of software metrics in contracts) and also the historical data of ISBSG. This was a short visit, only three days, but it made a deep impression on me. SPI China actually has the possibility to help the Chinese government to improve the estimation maturity in the software industry, and by doing so to improve the success rate of software projects and therefore avoid billions of wasted euros and wasted resources. SPI China can do this, and Nesma and ISBSG can help!
By the way, this is what Nesma already tries to do for years in the Dutch government, but unfortunately the Dutch government does not seem to be in the mode that they want to be helped yet. This is very strange, as according to research by the Dutch IT magazine Computable, 此更新将确保 Nesma FPA 标准的清晰度得到改进 Dutch government wastes 4 至 5 billion Euro per year on ICT projects that fail. Especially the large projects are failing often: of the projects over 7,5 million only 7% is successful. In total, 只要 30% are successful. Only 36% of all projects are being used after delivery. This is an outrageous waste of time, tax payer’s money and resources. Especially the cancelled projects, where valuable resources were wasted, while they could have worked on something useful instead. And in the end, a solution is probably still needed for the problem the project tried to solve. Many of the problems could definitely have been prevented by estimating the projects in a realistic way, by using mature estimation techniques. The following graph shows the estimation maturity model, developed by Dan Galorath.
As you might know, humans are hardwired to be optimistic. They usually don’t think in probabilities, but think of the most optimal way something can get done. Nobel prize winner Kahneman showed research where even people that were told that they were optimistic when estimating a certain task, actually still proved to be optimistic. 因此, estimates that are based on human (expert) 意见, experience, views, ideas and beliefs are very likely to be optimistic and need to be mitigated for this bias. Steve McConnell also shows this in his excellent book Software Estimation – demystifying the black art. In the Estimation maturity model, level 0 and level 1 are the stages where the human factor is not mitigated. This means that these estimates are very likely to be optimistic. Starting with an optimistic estimate is a huge red flag for any project, as it will start with a bad planning: a team that is too small and the duration to short. In the end these projects usually are much more expensive: non-linear extra costs as McConnell states in the same book. From Level 2 onwards, the effect of the human factor is mitigated by doing formal sizing as part of the estimate. Understanding the size of the software that needs to be developed greatly helps to estimate the effort needed and the duration needed, using relevant historical data (preferably company data, but ISBSG data is also helpful).
As the estimate is that 95% of the software projects in the industry is estimated using level 0 or level 1 实践, it’s easy to understand that many problems can be avoided just by improving the estimation maturity. SPI China and the Chinese government understand this! Some parts of the Dutch government also understand this, but for instance the relatively new ‘ICT测试室 (比特)’ seems unaware of Nesma or at least seems not willing to start the dialogue with Nesma. Maybe the fact that the Chinese government is taking Nesma seriously will be an eye-opener for them as well.
Anyway, I returned to China for the third time at the end of May, again to Beijing. This time a slightly longer visit so I could enjoy some nice trips to the Great Wall, the Olympic Parc and the Forbidden city. On Friday May 27, the conference was organized and I did not really what to expect. I had prepared my keynote speech with the title ‘Improve estimation maturity using functional size measurement and historical data’, basically covering the topic above. The presentation can be viewed on 幻灯片共享.
It turned out that SPI China had also some additional plans with me. In the morning, I was the middle point of a Q&A session with a number of high level executives of SPI China members, most of them government e.g. China Bank, China Post and the State Power Grid. I got to answer a lot of questions regarding the use of function points in agile projects, enhancement projects, outsourcing contracts and lots of other topics. Of course all questions were asked in Chinese, translated into English by deputy Secretary General of China SPI Mr. Ali Zheng, answered by me in English, and translated back into Chinese. So this took some time and maybe some of the points may have gone lost in translation, but everybody seemed to be happy and satisfied with the answers. It was also interesting for me, as I learned for instance that China Post already uses functional size measurement in a lot of their contracts with vendors and is very happy with this as it has reduced the chaos in contracting greatly.
Then after lunch the conference started. At some point during the conference I was asked to hand over a number of participation certificates to the people that had attended my courses in April. 当然, this was a nice ceremony with appropriate music playing.
At the end of the conference, I also was invited to take part in a discussion forum, where people from the audience could ask questions to a few people on the podium on topics related to function points and software estimation. This forum was very interesting and showed that there is a lot of potential in China for functional size measurement and Software Cost estimation.
This was all very exciting of course, but the main purpose of these visits was to arrange a partnership between Nesma and China SPI. We succeeded in drawing up an arrangement where SPI China and Nesma are now partners. SPI China has obtained the exclusive rights on the Chinese market to distribute Nesma products and services.
A number of concrete actions we are going to do are:
- Translate Nesma documents (including the measurement guidelines) and guides into Chinese and offer them for download on a special Chinese section on the website.
- The Nesma exams are going to be translated into Chinese.
- Nesma will organize a number of ‘Train the Trainer’ workshops to people that are eligible to participate according to both Nesma and SPI China. After the training and after passing the Nesma/Cito exam, the trainer will be accredited to be a Nesma trainer in China.
- The accredited trainers are going to train the functional size measurement specialists and cost estimators in China.
- Nesma will keep assisting SPI China in spreading the word by speaking at important events
此外, ISBSG and SPI China also renewed their partnership. In practice this also means that a number of ISBSG documents are going to be translated into Chinese. In addition SPI China and ISBSG are going to collect data in the Chinese market and make ISBSG data more accessible for the Chinese market.
I am very happy with these partnerships and I’m confident that we can have a number of materials available in Chinese before the big Chinese year 2017 starts!
Harold van Heeringen 是 Metri 的高级基准测试顾问. 除了他为 Metri 所做的工作, 他参与了 Nesma (董事会成员), 国际软件基准标准小组 (ISBSG, 现任总统) 和通用软件度量国际联盟 (COSMIC, 国际咨询委员会, 代表荷兰).