Someone normally has a bright idea or something needs improving. And then want a program to do it. But before hitting the keyboard there are some important precursors and it’s suprising how many companies and projects fail at this very first task with usually painful consequences. It’s called blast-off or initiation or … OK the name doesn’t matter but the aim is to setup the project on a good foundation and these principles aren’t specific to software development:-

Whats the aim? be clear about why we are doing this. Just because you can doesn’t mean you should. Where’s it fit? is this part of an overall strategy? And if so what is it? Are we sure we want this product? Timescales how much time have we got? Take too long and the product may become obsolete. Risks what are realistic risks and the consequences of failure? The benefits of success? Costs roughly. Millions or pennies? Timescales also connects with this - are there hard-deadlines and are they achievable? And do remember those opportunity costs![1] Who we need someone to be responsible to make it happen, often called a sponsor. This isn’t likely to be the actual techies who do the work… This appraisal really isn’t a big thing and shouldn’t take long even for big stuff. We can make it more formal and part of a methodology (eg Volere) or not. As the Roman politician Seneca said “If a man does not know to what port he is steering, no wind is favorable to him.” [2] This is all about being clear about what port you are sailing to and putting it in writing. - - - - - -

So, to take my own medicine: The aim is complete a realistic piece of software. I’m not considering commercial success or even critical acclaim so it won’t be a particularly novel or exciting game. The intent is to get into a habit of coding and finishing. A game because it’s more fun, fits into a broader hope to write indie games for income and readers will be able to understand it better. I’m reducing the risk of non-completion by making this a few months project, keeping it graphically simple and using familiar technology. - - - - - -

I’m not sure why sooooo many companies I’ve worked for have missed or skimped this first step. Sometimes the projects seems so small, or it’s all too rushed or nobody actually has answers and they don’t want to look stupid. They all rue the mistake ~sigh~ Tech-takeaway

Publicly ask such questions and get answers in writing. At the risk of seeming an nay-sayer it will save your arse later. At the end of a project (death or success!) reflect on how good the starting appraisal was.
[1] opportunity cost - the lost benefit or income from the things you won’t be able to do because you are tied up with plan A. [2] http://www.quotationspage.com/quote/28685.html