Ten wastes of software research and development: the other side of research and development efficiency

“ Two dozen programmers ,3 Year time ,4732 individual bugs, And the relentless pursuit of extraordinary software ”

《 Dreambreak code 》 This book , I saw it ten years ago , Read it in one breath . I was still Cisco（ Cisco ） Work , I feel that the R & D team has made mistakes , In this book, you can basically see .

That year Lotus 1-2-3 Designer Mitchell Kapor, Leave Lotus After that, the open source application foundation was established (OSAF), Recruited a group of great programmers , Develop the so-called revolutionary next generation personal information management system －－Chandler. This team seems to have no shortage of money 、 There is no lack of technology 、 No lack of experience , But we can't release a seemingly simple software .6 Years have passed ,Chandler Reluctantly struggled to release 0.7 edition , Spend millions of dollars , But it failed in the end , Dreambreak code .

In a 6 In the marathon of , Made a series of mistakes , As a result, the promising project was half dead . for example ,Chandler The project members decided to use P2P Architecture to share calendars , But no effort has been made to design relevant algorithms or protocols , Instead, spend a lot of time discussing Chandler The interface of , This drag is a few months , Last P2P Architecture is completely abandoned , It's a great waste . Do something like this , Then it was overthrown , It is also common in software development .

The last one discussed   Negative list of software R & D Efficiency ： Which is the number one enemy ？ today , Discuss the other side of software development effectiveness , The opposite of software effectiveness , Causing low software efficiency “ waste ”. Waste can be divided into ：

• Direct waste ： Unnecessary development costs or directly perceived waste , Such as （ Code that nobody uses ; Commented out code , Functions that have never been used 、 Over testing, etc ;

• Indirect waste ： Not directly visible development costs , Such as inferior quality 、 Code complexity 、 The extra cost of inefficient communication .

It doesn't matter if it's a direct cost , Or indirect costs , Don't look at the cause of it , But to see the results it produces , I summarize it as ten wastes .

The first 10 A great waste ： The work done did not bring benefits into play in time

for example , The finished code is not submitted to the code base in time to build , In the developer's local machine ; The tested functions have not been delivered to users yet , Equivalent to in the company's inventory , There is no benefit .

The first 9 A great waste ： Switching between different roles or tasks

for example , One person participates in multiple projects , Need to switch between different tasks , Familiar with too much business and project background , Often switch thinking 、 Switch virtual workspaces , It will cause more time waste .

The first 8 A great waste ： Unnecessary handover in software

Handover of tasks in software , Naturally, it will increase the learning cost , Increases the cost of communication , Although there is inevitable handover in daily software development , But unnecessary or excessive handover , Will bring waste , for example ：

• The personnel turnover rate is relatively large （ If higher than 20%）, Work handover between the old and the new

• The transition between developers and testers , Such as development 、 Testing is two relatively independent teams ;

• Handover of software from development to deployment （ If effective DevOps, This kind of cost is very low ）

The first 7 A great waste ： Overwork （ Out of range ）

Because of the lack of an effective process 、 Documentation and consistency requirements, etc , The scope of the project or task is not accurately understood in the work , Did work beyond the scope , Or can't do just the right thing , Including over management 、 Over testing 、 Write too many documents 、 Over communication （ Too many meetings ） etc. .

The first 6 A great waste ： wait for

Many people wait for others to work , For example, the communication and cooperation of the team are not active , Wait for others to come to the door ; There is no tacit cooperation between development and testing , The test waits for the development to submit a new version ; The links are not well connected , There will be waiting in the middle . Even in a team or a job, there is waiting , For example, continuous testing is not achieved , There will be waiting . Sometimes the test environment is not ready , Can't test , There will also be waiting .

The first 5 A great waste ： Make mistakes again and again

No root cause analysis , Headache doctor 、 Foot pain cures the foot . Some people in the team don't make mistakes , But others do it again ; Some teams don't make mistakes , But other teams do it again ; Mistakes can be all kinds of , Including bad planning 、 Wrong document version, etc .

The first 4 A great waste ： Repeat the wheel

There are open source tools or mature commercial tools on the market , It is not used directly , Or buy tools directly , Instead, you develop it yourself .

The first 3 A great waste ： rework

Due to the lack of unified specifications 、 Complex system 、 Weak personnel ability , Lead to inferior quality 、 There are many defects , Cause redesign 、 Rewrite code , It's all rework . Excessive code refactoring 、 Regression testing and so on are all classified as waste caused by rework .

The first 2 A great waste ： Useless functions and code  

Useless functions and code , similar “ overproduction ”, According to some statistical results , In existing software applications , As many as 2/3 Features are rarely or never used . Including code that nobody uses 、 Commented out code, etc .

The first 1 A great waste ： The direction of the whole product is wrong  

For user needs 、 Wrong direction of business understanding , The whole product failed after launching , Nobody uses .

Reference resources ：

• The underlying logic of software R & D efficiency

• Hit the soul ： The first principles of software development and 10 Double efficiency

• Negative list of software R & D Efficiency ： Which is the number one enemy ？

   

Who is the number one enemy of software development ？

Welcome to click   The questionnaire system   Take part in the questionnaire , Thank you very much