Many developers have plenty of skill, industry, and good ideas but want the ability to finish projects. What follows is a project strategy that can help.

PHASE ONE: Plan
1. Scope out what the project is about. This is whatever you can think of.
2. Cut scope down to something absurdly basic and modest. If this does not represent a D- grade, then the scope needs to be cut down more. This is what you are committed to delivering.
3. Set a deadline for how long you will spend on the project as a whole. This deadline will not move. The product is whatever the project has produced by that deadline.

PHASE TWO: Pass
Focus efforts only on hitting the basic commitment. Once you are done, your project has a passing grade (albeit a poor one).

PHASE THREE: Perform
Now look over what work is left and how much time is left. Choose to do the work that gets the best results for the time spent. You want to bring the grade of your project up as high as possible. When time is up, you are done and the project is done.

Why is it hard to finish software projects? Let us first look at what that means. If you do a software project that means you form up an intention to write some software and you carry out that intention. When the software is finished, the project is finished. When the project is finished, the software is finished. You do the project and you make the software. Why is that so hard?

It is hard because you do not really know what you are doing. The nature of software is such that you cannot really know. Software is always changing, always expanding out. It is doing this even while you are working on it. It is doing this because you are working on it. This why projects go on and on and you wonder what you have to do to make it be finished.

Suppose there is a solution. What does that mean? You are able to know what you are going to do and how you are going to do it. You have something that you can follow through on and get done. Yet somehow at the same time you have to also be able to not know what you are doing or how you are going to do it.

The solution is this: Make the project be to finish something basic you know about and then continue working up to a deadline. Make the software be that basic minimum plus something else. What else? That is down to how you spend the time up to the deadline. It will be how good you can make it.

And suppose the project ends and there is more work to be done? There can always be more projects for you to do. You can build out your program over time.

• ? Can this work in a corporate environment?

The strategy also played into writing this piece.