Home >Backend Development >C#.Net Tutorial >Avoid six common development mistakes
Tony Patton posted an article on ZDNet titled "Avoid Six Common Development Mistakes". I thought it was good after reading it, so I shortened it and posted it.
1. Think you must use Visual Studio.NET to develop .NET applications
You only need to install the .NET framework which can be downloaded for free on the Microsoft website. With the .NET Framework installed, you can develop using your favorite text editor and .NET tool command line. Moreover, you can also use third-party development tools, such as SharpDevelop.
2. Exclusive Information
One of the defining characteristics of developers is their tendency to keep information secret. Knowing abilities that others don't know often leads to a false sense of superiority. In contrast, sharing information helps the team expand its knowledge. Plus, the information shared can be improved as other team members join the team. Finally, your knowledge is expanded when you explain an idea or concept to another developer or group of developers.
Teamwork is the driving force behind movements like Agile Software Development. In addition, many teams utilize peer code reviews, in which developers review and comment on other developers' code, which promotes learning and understanding on both sides.
3. Ignore code comments
Annotating your own code is a simple and direct method, but why do people always ignore this? Some developers mistakenly believe that reducing documentation (of any kind) will improve job security. This is the same fallacy as the exclusive information discussed earlier.
A small number of well-arranged comments can greatly improve the readability of the code and facilitate code management. There is a tendency to overuse annotations. Basically, these comments should give a brief description of the code, variables, methods, etc. Many teams like to insert comments when making changes to their code. When (if) something goes wrong, it's easy to find out what changes were made in the past.
4. Not seeing the value of a version control system
Most developers are very familiar with this situation: you launch a new version of an important web application, and early the next morning your phone keeps ringing with all the Site problem cry.
As long as you are willing to admit that you may have made a mistake or two, a version control system can be your best friend in that situation. By using version control systems such as CVS, IBM Rational ClearCase, and Microsoft Visual SourceSafe, you can easily revert to previous versions of your application for production. This gives you time to identify the problem in the development environment while the production site is ready. This approach is far superior to hunting through production code to find the problem.
Also, the source control system allows different team members to develop the same code at the same time. When developers review modifications to the system, these changes are combined.
One of the reasons some developers dislike version control systems is because it adds a step to the development process, which means the process of committing code to the source control system and restoring it through it is slower. This extra wait has dampened the patience of some developers.
5. Thinking that all projects require network services
Due to the over-hyping of network services, many developers believe that network services must be used in some form when developing projects. Although it may be necessary or appropriate to use web services during development, it is not always possible. You need to analyze different situations to decide whether it is appropriate to perform network services. For example, another technology such as remoting may be more appropriate.
6. Ignoring the importance of confirming data
In application code, data is processed in different forms. For this reason, developers should confirm the integrity and validity of data during the development process. This means that the data is read through a terminal database that returns zero characters. The data is checked before execution because zero values cannot be used in most operations.