Do you want to create JPA repositories without writing any boilerplate code? If so, read my Spring Data JPA tutorial!

There Are No Small Projects

Small plants

How often have your heard one of the following phrases:

This is a small project so we will just have to put something together and FAST.

Big projects need to be designed in a totally different manner than small ones.

This is just a campaign product / prototype which is used only once. We can write tests later if the scope of the project is increased.

These phrases and many others like them suggest that big software projects require different approach than small ones. Even though this is partially true, each one of those phrases contains a hidden agenda:

First, they propose that small projects should be put together as fast as possible. Second, they propose that good software development practices such as unit testing, integration testing and code reviews are not as important in a small project than they are in a big one.

I have noticed that this attitude is surprisingly common but that does not make it right. I personally believe that this approach is flawed for the following reasons:

Not all projects start as big ones. In these cases the scope of the project is extended during the project. If you have separate practices for small and big projects, you will end up noticing that you have been following the wrong practices. This can be a critical mistake since you might have to go back to the code you have written earlier and adapt it to match the demands of the big projects This costs both time and money and your customer is often not willing to pay for it. And guess what? The customer is absolutely right. You made the mistake. You will have to pay the prize.

Small projects are important to your customer. It does not really matter whether the project is described to you as a temporary or as a permanent solution. The fact is that you cannot know if something you delivered is going to be a part of something bigger in the future. The only ethical thing to do is to ensure that you finish all projects by following good software development practices. This way it should be easier to extend the software you delivered if needed.

Remember that even if the project does not sound big and important project to you, it might be a huge investment to your customer and it should be treated as such. If you can ship the finished product with your name and contact details, you have most likely given the customer the respect he deserves.

Changing peoples attitude is hard. If your developers have learned that it is acceptable to cut corners in small projects, you are going to have a hard time to change their attitude when you finally get that big project you have been dying to get. The reason why this is hard is that you will have force your developers out of their comfort zones and people tend to resist changes like this. That is why the best thing you can do for your business is to follow good software development practices in every project. It will save both your nerves and your money.

Keeping your expectations high has also got a one additional benefit: It will make the recruitment process much easier for you. As long as you mention your high expectations during job interviews, it will be fairly easy to see who is not a good addition to your team.

I have now given you three reasons why you should treat each project in the same way. Remember that it is perfectly fine to decline projects which seem too small for you. However, if you decide take a project (either big or small), you should always follow the same principles and aim for delivering software which you can be proud of. If you are a real professional, anything less than this is simply unacceptable.

If you enjoyed reading this blog post, you should follow me on Twitter:

About the Author

Petri Kainulainen is passionate about software development and continuous improvement. He is specialized in software development with the Spring Framework and is the author of Spring Data book.

About Petri Kainulainen →

0 comments… add one

Leave a Comment