Hi there ,
i want to start a new project and i don't know where to start.
After learning the last 2 projects in teams, i learned new techniques to plan a project. Like "the planning game". OR should i make use cases, although i won't look at it after creation.
But now i am coding for my private project. So, do i need planning at all ? Or do you think it is a waste of time ?
It totally depends on the project, and your own style. Writing some informal use cases and posting them on the wall can keep you aware of your targets whilst coding. Breaking up the project into smaller projects can help you to set deadlines and keep a schedule. Just to name two things that may or may not be relevant.
Too much overheard is too much overhead. But you should consider at least following a process. I think it's not a question of whether to plan, but which process is most efficient to support your endeavor.
Consider the agile process. If this is a project you are doing alone, you would effectively be the developer, manager, and the client.
At minimum, I would brainstorm, produce a schedule, and a design document that explains the core classes that comprise your product.
I find it very useful to do some planning for solo projects. If this project is being done in a job environment, there may be a chance that business stake holders will need an idea of what is being delivered and when. If it is for a personal project, even in this case planning can be very useful. While I am not a believer in "big design up front," having some sort of a flexible roadmap and design helps me to focus.
I try to itemize the concrete tasks that need to be done, have some sort of a database schema if applicable and a rough class diagram. I don't necessarily have a detailed UML, but some kind of scaffolding is very helpful.
As I "answer" the more tough design questions, I make sure to document these. This way I don't for get them and have to waste time thinking about them again. Just keep in mind that everything can and may change.
Unless it's so trivial you can keep it all in your head, then yes, you do need some sort of planning effort IMO. If this is a private project and you're employed, you'll likely be working on it in spurts. How do you figure to stay on track between work sessions without a plan? What if you can't do anything for a month? Will you be able to remember everything & restart without re-planning?
Use a process you're comfortable with, but you need to have some sort of plan or you'll spend more time thrashing than working.
For all of my 1-man projects, I start with a rough outline of tasks and assign rough effort estimates. As I get in to working on each task, I refine it by creating subtasks with more detailed estimates.
I use the free version of FogBugz (it's free for up to 2-man teams). It integrates right into my IDE (VS or Eclipse) so I can refine my project plan as I go, minimizing the overhead while leaving me with a good sense of what I will have done by when.
No plan = no clue what you are doing. A trip without a map is going to go nowhere fast.
You need some level of planning just to map out what you are trying to achieve.
If you are experienced in project planning you may be able to whip a plan up in a few minutes inside your head, but it always helps to get it down on paper. If you are not experienced then you need to take more time and elaborate out your goals and how you plan to reach them.
There is no need to go overboard with it of course, so no formal documents or flow-charts; that level of planning would be going to far. But you must plan.
A plan will help keep you focused and will help you avoid mistakes.