How to Start a Software Project as Fast as Possible
We started a new project last week, and I started to wonder what are the most important things to do if we want to get a project started as fast as possible (without messing it up).
I figured out three things which have helped me to concentrate on the right things during the early phases of a software project.
These things are:
1. Don’t Aim for Perfection
When a software project is started, the first thing that we have to do is to agree on the practices used in the project.
I have noticed that the most efficient way of doing this is to follow these steps:
- Create a To-Do list which contains the things which must be taken into account. Typically this list contains things like version control system, project model, technologies, documentation, and so on.
- Invite the project team to a meeting and go through your to-do list one by one. Try to decide as many things as you can on this meeting.
- If you cannot decide something or a decision creates a need to do something, create a task, assign it to a team member (shared responsibility is nobody’s responsibility), and set a deadline.
- Don’t forget to follow up!
The idea is to get everything done as fast as possible, and use retrospective meetings to drive continuous improvement.
2. Use a Project Template
Most projects which use the same technology stack have a common foundation. This foundation includes the build process, dependencies, configuration, and some common functions (registration, login, and so on).
Creating this foundation from scratch every time when it is required makes no sense. It is a waste of time and money.
We can solve this problem by creating a project template or using an existing one. A good project template has one goal:
It ensures that we don’t have to spend any time for doing boring grunt work!
This means that we can concentrate doing work that matters.
You can find many project templates from the internet. If you are working with Spring web applications, you might want to take a look at the following projects:
- JHipster is a Yeoman generator which is used to create Maven + Spring + AngularJS projects.
- SpringFuse is an online service which generates code for data-oriented Java applications.
3. Don’t Worry About Features (Yet)
When we start a new project, we should not care about features of the application. We should care only about these two goals:
- Decide the common practices used in the project.
- Create a solid foundation which is easy to extend.
If we worry about the features of the application when we are starting a project, we spend a lot of energy on worrying about things which aren’t essential right now. This will decrease our chances to meet our goals.
Don’t get me wrong. Solving the right problem is crucial to our success but right now the most important thing that we have to worry about is to get our project started.
There Is a Time and Place for Everything
I think that the easiest (and often the best) way to increase our efficiency is to concentrate on the right thing at the right time. This seems obvious but the next question is a lot harder to answer:
What is the right thing?
I provided one answer to that question. Is my answer the right one?
I have no idea. Ask me again after a few years and I will let you know.
P.S. If you think that I am wrong or I missed something, share your opinion on the comment section.
Good